CN112905026A - Method, device, storage medium and computer equipment for displaying word suggestions - Google Patents

Method, device, storage medium and computer equipment for displaying word suggestions Download PDF

Info

Publication number
CN112905026A
CN112905026A CN202110343650.6A CN202110343650A CN112905026A CN 112905026 A CN112905026 A CN 112905026A CN 202110343650 A CN202110343650 A CN 202110343650A CN 112905026 A CN112905026 A CN 112905026A
Authority
CN
China
Prior art keywords
word
character
character string
words
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110343650.6A
Other languages
Chinese (zh)
Other versions
CN112905026B (en
Inventor
夏文兵
王元阳
杨思
高嵩
张仕翼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Perfect World Holding Group Ltd
Original Assignee
Perfect World Holding Group Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Perfect World Holding Group Ltd filed Critical Perfect World Holding Group Ltd
Priority to CN202110343650.6A priority Critical patent/CN112905026B/en
Publication of CN112905026A publication Critical patent/CN112905026A/en
Application granted granted Critical
Publication of CN112905026B publication Critical patent/CN112905026B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233—Character input methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/34—Browsing; Visualisation therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/10—Text processing
    • G06F40/166—Editing, e.g. inserting or deleting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/237—Lexical tools
    • G06F40/242—Dictionaries

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Human Computer Interaction (AREA)
  • Document Processing Apparatus (AREA)
  • Machine Translation (AREA)

Abstract

The invention discloses a method, a device, a storage medium and computer equipment for displaying word suggestions, relates to the technical field of information, and mainly aims to improve word recommendation precision in a word searching process and ensure that the word suggestions can meet the searching requirements of users. The method comprises the following steps: acquiring a character string input by a user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; determining relative position information of the first different character pair on a character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; and displaying the words meeting the preset conditions as word suggestions based on the editing distance. The invention is mainly suitable for displaying word suggestions.

Description

Method, device, storage medium and computer equipment for displaying word suggestions
Technical Field
The present invention relates to the field of information technologies, and in particular, to a method, an apparatus, a storage medium, and a computer device for displaying word suggestions.
Background
Because english education is more important at present society, has raised english study's heat tide, various english study class APPs on traditional computer, panel computer and the smart mobile phone are also more and more, and the user often can utilize various APPs to search for the word when studying english.
At present, in the process of word search, fuzzy matching is usually performed on a character string input by a user and a character string contained in each word in a dictionary, and word information is recommended for the user according to a fuzzy matching result. However, in this way, once a certain character input by the user is inconsistent with a corresponding character in a word, it is considered that there is no association between the two characters, but it is likely to be caused by a wrong click of the user on the keyboard, so the recommendation information provided by this matching way has a large deviation, the accuracy of information recommendation is low, the recommendation effect is not good, and the user experience is poor.
Disclosure of Invention
The invention provides a method, a device, a storage medium and computer equipment for displaying word suggestions, which mainly aim to improve word recommendation precision in a word searching process and ensure that the word suggestions can meet the searching requirements of users.
According to a first aspect of the present invention, there is provided a method of presenting word suggestions, comprising:
acquiring a character string input by a user;
matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair;
determining relative position information of the first different character pair on a character input keyboard;
calculating the editing distance between the character string and each word according to the relative position information;
and displaying the words meeting the preset conditions as word suggestions based on the editing distance.
Optionally, the matching a first character in the character string with a second character in each word in a preset lexicon to obtain a first different character pair includes:
judging whether the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word in each word:
if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, matching a first character in the character string with a second character in the target word to obtain a first different character pair;
and if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, inserting or deleting the target characters in the character string, and matching a first character in the character string after the target characters are inserted or deleted with a second character in the target word to obtain a first different character pair.
Optionally, the inserting or deleting a target character in the character string includes:
comparing the character string with the target word, and determining the target character inserted or deleted in the character string according to a comparison result;
and inserting or deleting the target character in the character string, so that the similarity between the character string after the target character is inserted or deleted and the target word is larger than a preset threshold value.
Optionally, the calculating, according to the relative position information, an edit distance between the character string and each word includes:
if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, determining a distance value corresponding to the first different character pair according to the relative position information;
and determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair.
Optionally, the calculating, according to the relative position information, an edit distance between the character string and each word includes:
if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, determining a first editing distance corresponding to the character string according to the number of the characters corresponding to the target character;
determining a second editing distance corresponding to the character string according to the relative position information;
and adding the first editing distance and the second editing distance to obtain the editing distance between the character string and the target word.
Optionally, after obtaining the distance value corresponding to the first different character pair according to the relative position information, the method further includes:
determining the correct rate weight corresponding to the first different character pair according to the position information of the first character in the character string; and/or
Inquiring a preset historical error weight table according to the first different character pair, and determining the historical error weight corresponding to the first different character pair;
determining an editing distance between the character string and the target word according to the distance value corresponding to the first different character pair, including:
determining an edit distance between the character string and the target word based on the correct rate weight and/or the historical error weight, and the distance value.
Optionally, the displaying, based on the edit distance, a word meeting a preset condition as a word suggestion includes:
judging whether the character string is a word prefix in the preset word bank;
if the character string is a word prefix, correcting the editing distance between the character string and each word containing the word prefix to be 0;
and sequencing the words from small to large based on the corrected editing distance, and displaying the words meeting the preset conditions as word suggestions.
Optionally, the method further comprises:
when the first different character pair is matched, prompting a user to perform voice input;
determining words corresponding to the voice information input by the user, and adding the words corresponding to the voice information into the word suggestions for displaying, wherein the display sequence of the words corresponding to the voice information in the word suggestions is in a preset front display range.
Optionally, the displaying, based on the edit distance, a word meeting a preset condition as a word suggestion includes:
taking the words with the editing distance smaller than the preset editing distance in the words as word suggestions;
determining the display priority corresponding to the word in the word suggestion according to the editing distance corresponding to the word in the word suggestion;
and displaying the words in the word suggestions based on the display priority.
Optionally, the presenting the words in the word suggestions based on the presentation priority includes:
determining words matched with the user learning task in the word suggestions, and correcting the display priority corresponding to the words matched with the user learning task;
and displaying the words in the word suggestions based on the corrected display priority.
Optionally, the presenting the words in the word suggestions based on the presentation priority includes:
if the user history clicking record corresponding to the character string exists, determining words existing in the user history clicking record in the word suggestion;
according to the click frequency and/or click time corresponding to the words existing in the user history click record in the word suggestions, correcting the display priority corresponding to the words existing in the user history click record in the word suggestions;
and displaying the words in the word suggestions based on the corrected display priority.
Optionally, the presenting the words in the word suggestions based on the presentation priority includes:
if the user history clicking record corresponding to the character string does not exist, collecting history clicking records of other users, and sequencing the clicked words according to clicking frequencies corresponding to the clicked words in the history clicking records of the other users;
determining target point word selection with the ranking rank within a preset range according to the ranking result;
judging whether the target point word selection exists in the word suggestions or not;
if the target point word selection exists in the word suggestions, correcting the display priority corresponding to the target point word selection;
and displaying the words in the word suggestions based on the corrected display priority.
Optionally, after the determining whether the target point vote exists in the word suggestion, the method further includes:
and if the target point vote does not exist in the word suggestions, adding the target point vote meeting the preset top ranking condition into the word suggestions, and displaying the words in the added word suggestions.
Optionally, the presenting the words in the word suggestions based on the presentation priority includes:
acquiring learning state information of word marks in the word suggestions;
based on the learning state information, correcting the display priority corresponding to the words in the word suggestions;
and displaying the words in the word suggestions based on the corrected display priority.
According to a second aspect of the present invention, there is provided an apparatus for presenting word suggestions, comprising:
the device comprises an acquisition unit, a display unit and a control unit, wherein the acquisition unit is used for acquiring a character string input by a user;
the matching unit is used for matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair;
a determining unit for determining relative position information of the first different character pair on a character input keyboard;
a calculation unit configured to calculate an edit distance between the character string and each word according to the relative position information;
and the display unit is used for displaying the words meeting the preset conditions as word suggestions based on the editing distance.
Optionally, the matching unit includes: a judging module, a matching module and an inserting module,
the judging module is used for judging whether the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word in each word;
the matching module is used for matching a first character in the character string with a second character in the target word to obtain a first different character pair if the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word;
the inserting module is used for inserting or deleting the target character in the character string if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, and matching a first character in the character string after the target character is inserted or deleted with a second character in the target word to obtain a first different character pair.
Optionally, the insertion module includes: a comparison sub-module and an insertion sub-module,
the comparison submodule is used for comparing the character string with the target word and determining the target character inserted or deleted in the character string according to a comparison result;
the inserting sub-module is used for inserting or deleting the target character in the character string, so that the similarity between the character string after the target character is inserted or deleted and the target word is larger than a preset threshold value.
Optionally, the computing unit includes: a first determination module and a second determination module,
the first determining module is configured to determine, according to the relative position information, a distance value corresponding to the first different character pair if the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word;
and the second determining module is used for determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair.
Optionally, the computing unit includes: a first determining module, a second determining module and an adding module,
the first determining module is configured to determine a first editing distance corresponding to the character string according to the number of characters corresponding to the target character if the number of characters corresponding to the character string is different from the number of characters corresponding to the target word;
the second determining module is configured to determine a second editing distance corresponding to the character string according to the relative position information;
and the adding module is used for adding the first editing distance and the second editing distance to obtain the editing distance between the character string and the target word.
Optionally, the first determining module is further configured to determine, according to the position information of the first character in the character string, a correct rate weight corresponding to the first different character pair; and/or querying a preset historical error weight table according to the first different character pair, and determining the historical error weight corresponding to the first different character pair;
the second determining module is specifically configured to determine an edit distance between the character string and the target word based on the correct rate weight and/or the historical error weight, and the distance value.
Optionally, the display unit comprises: a judging module, a correcting module and a display module,
the judging module is used for judging whether the character string is a word prefix in the preset word bank;
the correction module is configured to correct the edit distance between the character string and each word including a word prefix to 0 if the character string is the word prefix;
and the display module is used for sequencing the words from small to large based on the corrected editing distance and displaying the words meeting the preset conditions as word suggestions.
Optionally, the apparatus further comprises: a prompting unit for prompting the user to perform a prompt operation,
the prompting unit is used for prompting a user to perform voice input when the first different character pair is matched;
the display unit is further configured to determine a word corresponding to the voice information input by the user, and add the word corresponding to the voice information into the word suggestion for display, where a display sequence of the word corresponding to the voice information in the word suggestion is in a preset front display range.
Optionally, the display unit comprises: a determining module and a display module, wherein the display module is used for displaying the display information,
the determining module is used for taking the words of which the editing distance is smaller than the preset editing distance in the words as word suggestions;
the determining module is further configured to determine a display priority corresponding to a word in the word suggestion according to an editing distance corresponding to the word in the word suggestion;
and the display module is used for displaying the words in the word suggestions based on the display priority.
Optionally, the display module includes: a correction submodule and a display submodule,
the correction submodule is used for determining words matched with the user learning task in the word suggestions and correcting the display priority corresponding to the words matched with the user learning task;
and the display submodule is used for displaying the words in the word suggestions based on the corrected display priority.
Optionally, the display module includes: a determination submodule, a correction submodule and a display submodule,
the determining submodule is used for determining a word in the user history clicking record in the word suggestion if the user history clicking record corresponding to the character string exists;
the correction submodule is used for correcting the display priority corresponding to the word existing in the user history clicking record in the word suggestion according to the clicking frequency and/or the clicking time corresponding to the word existing in the user history clicking record in the word suggestion;
and the display submodule is used for displaying the words in the word suggestions based on the corrected display priority.
Optionally, the display module includes: a sequencing submodule, a determining submodule, a judging submodule, a correcting submodule and a display submodule,
the sorting submodule is used for collecting historical click records of other users if the user historical click record corresponding to the character string does not exist, and sorting the click words according to the click frequency corresponding to the click words in the historical click records of the other users;
the determining submodule is used for determining target point word selection with the ranking rank within a preset range according to the ranking result;
the judgment submodule is used for judging whether the target point word selection exists in the word suggestions or not;
the correction submodule is used for correcting the display priority corresponding to the target point word selection if the target point word selection exists in the word suggestion;
and the display submodule is used for displaying the words in the word suggestions based on the corrected display priority.
Optionally, the display module further includes: a sub-module is added to the system,
and the adding submodule is used for adding the target point vote meeting the condition of the preset rank before to the word suggestion if the target point vote does not exist in the word suggestion, and displaying the word in the added word suggestion.
Optionally, the display module includes: an acquisition submodule, a correction submodule and a display submodule,
the obtaining submodule is used for obtaining the learning state information of the word marks in the word suggestions;
the correction submodule is used for correcting the display priority corresponding to the word in the word suggestion based on the learning state information;
and the display submodule is used for displaying the words in the word suggestions based on the corrected display priority.
According to a third aspect of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above-described information search method.
According to a fourth aspect of the present invention, there is provided a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the above information search method when executing the program.
Compared with the mode that once the input characters are inconsistent with the corresponding characters in the words at present, no association exists between the two characters, the method, the device, the storage medium and the computer equipment for displaying the word suggestions can acquire the character strings input by the user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; at the same time, determining relative position information of the first different character pair on the character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; finally, words meeting preset conditions are displayed as word suggestions based on the editing distance, the editing distance between the character string and the words is calculated according to the relative position information of the first different character pairs on the character input keyboard, the layout factors of the character input keyboard can be referred in the calculation process of the editing distance, the calculation precision of the editing distance between the character string and the words is improved, the words meeting the preset conditions are recommended to a user as the word suggestions according to the editing distance, the word recommendation precision in the word searching process can be improved, the word suggestions can meet the searching requirements of the user, and the user experience is enhanced.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of a method for presenting word suggestions according to an embodiment of the present invention;
FIG. 2 is a flow diagram illustrating another method for presenting word suggestions provided by embodiments of the present invention;
FIG. 3 is a diagram illustrating a keyboard layout provided by an embodiment of the present invention;
FIG. 4 is a block diagram of an apparatus for presenting word suggestions according to an embodiment of the present invention;
FIG. 5 is a block diagram of another apparatus for presenting word suggestions according to an embodiment of the present invention;
fig. 6 shows a physical structure diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
At present, once a certain character input by a user is inconsistent with a corresponding character in a word, the two characters are considered to be not associated, but are probably caused by a wrong click of the user on a keyboard, so that the recommended information provided by the matching mode has large deviation, the precision of information recommendation is low, the recommendation effect is not good, and the user experience is poor.
In order to solve the above problem, an embodiment of the present invention provides a method for presenting word suggestions, as shown in fig. 1, the method including:
101. and acquiring the character string input by the user.
The method includes that a character string is an English character string currently input in a search box (or an input box) by a user, the character string can be specifically a prefix of a word, can also be a complete word, and can also not be a complete word and only contains partial characters. The embodiment of the invention is mainly suitable for scenes such as word searching, word inputting and the like, and the execution main body of the embodiment of the invention is a device or equipment capable of searching information or inputting information, and can be specifically arranged at one side of a client or a server.
102. And matching the first character in the character string with the second character in each word in a preset word bank to obtain a first different character pair.
Wherein the first character is a character in the character string, the second character is a character in the word, the first character in the character string is different from the second character in the word, when the first character in the character string and the second character in the word are in the same position order, the first character and the second character form a first different character pair, for the embodiment of the invention, if the character input by the user is different from the character in the same position order in the word, the character input by the user is determined as the first character, the character in the same position order in the word is determined as the second character, the first character and the second character form a first different character pair, namely when the user mistakas one character, the first different character pair is generated, during the character input by the user, a plurality of wrong characters may exist, namely when the third character in the character string and the fourth character in the same position order in the word are different, a second different character pair is generated, for example, the character string input by the user is "lukcy", the word in the preset lexicon is "lucky", the first character "k" in the character string and the second character "c" in the word form the first different character pair, the third character "c" in the character string and the fourth character "k" in the word form the second different character pair, and so on, a plurality of different character pairs can be formed, such as the third different character pair and the fourth different character pair.
103. Determining relative position information of the first different character pair on a character entry keyboard.
In the keyboard layout shown in fig. 3, when the character string input by the user is a leak, the relative position information of the first character "k" and the second character "m" in the word loam on the keyboard is diagonally adjacent, and the relative position information of the first character "k" and the second character "m" in the word loam on the keyboard is not adjacent; when the character input by the user is loan, the relative position information of the first character "n" and the second character "m" in the word loam on the keyboard is left-right adjacent. In addition, a large number of english words, specifically, an english dictionary, are stored in the preset word bank.
In the prior art, when the character string is matched with the words in the preset lexicon, once the characters in the character string are inconsistent with the corresponding characters in the words, no association exists between the characters and the corresponding characters in the words, but the inconsistency of the characters is probably caused by character input errors of a user due to local factors of a keyboard.
Specifically, when determining the relative position information of a first different character pair on the character input keyboard, firstly comparing the characters in the character string with the characters at the corresponding positions of any word in a preset word stock, and if the characters in the character string are the same as the characters at the corresponding positions in the word, not considering the relative position information of the characters on the character input device; if the characters in the character string are different from the characters at the corresponding positions in the word, a first different character pair is generated, the relative position information of the first different character pair on the character input keyboard needs to be determined, and the editing distance between the character string and the word is calculated according to the relative position information. For example, when the currently input character string of the user is lock, and the character string input by the user is matched with the word loam in the preset thesaurus, each character in the character string needs to be compared with the character at the corresponding position in the word loam, it can be determined by comparison that the first character "c" in the string lock is different from the second character "a" in the word "loam", the third character "k" is different from the fourth character "m" in the word "loam", thereby generating a first different character pair and a second different character pair, and determining relative position information of the first different character pair and the second different character on the character input keyboard, respectively, by referring to the layout of the keyboard, it can be determined that the relative position information between the first character "c" in the character string lock and the second character "a" in the word loam is not adjacent, and the relative position information between the third character "k" in the character string lock and the fourth character "m" in the word loam is diagonally adjacent.
Further, when referring to the layout factors of the keyboard, a preset keyboard character position information table may be queried in which left and right adjacent characters corresponding to each character in the keyboard and diagonally adjacent characters are recorded, for example, in which left and right adjacent characters corresponding to a character a including s, diagonally adjacent characters corresponding to a character a including q, w and z, left and right adjacent characters corresponding to a character z including x, and diagonally adjacent characters corresponding to a character z including a and s are recorded, whereby when determining the relative position information, the relative position information of a first different character pair on the keyboard may be determined by querying the preset keyboard character position information table, for example, when the character string input by the user is loak, the word in the preset lexicon is loam, when determining the relative position information between the first character "k" in the character string loak and the second character "m" in the word loam, by inquiring a preset keyboard character position information table, it is determined that left and right adjacent characters corresponding to a first character k comprise j and l, and diagonally adjacent characters corresponding to the first character k comprise i, o and m, so that the relative position information between the first character "k" in the character string loak and the second character "m" in the word loam can be determined to be diagonally adjacent.
104. And calculating the editing distance between the character string and each word according to the relative position information.
Wherein, the edit distance between the character string and the word refers to how many times the character string needs to be converted into the character string corresponding to the word through insertion, deletion and replacement operations, for example, the character string currently input by the user is loak, when the character string input by the user is matched with the word loam in the preset thesaurus, the prior art directly determines that the first character "k" is not associated with the second character "m" in the word loam, thereby causing the calculation precision of the edit distance to be low, the word information recommended by the user has larger deviation according to the edit distance, and the search requirement of the user cannot be met, but in the process of calculating the edit distance, the layout factor of the character input keyboard is introduced, the edit distance is calculated according to the relative position information of the first different character pairs on the input keyboard, thereby improving the calculation precision of the edit distance, the word information recommended to the user according to the editing distance can meet the searching requirement of the user, and the word searching precision is improved.
Specifically, when the edit distance between the character string and each word is calculated according to the relative position information, it is necessary to determine a distance value corresponding to a first different character pair according to the relative position information, and if there are a plurality of different character pairs, add the distance values corresponding to the plurality of different character pairs to obtain the edit distance between the character string and the word. The closer the distance between a first character in the character string and a second character at a corresponding position in a word on the character input keyboard is, the smaller the distance value corresponding to a first different character pair is, and the smaller the editing distance between the character string and the word in the preset lexicon is calculated; and conversely, the farther the distance between the first character in the character string and the second character at the corresponding position in the word on the character input keyboard is, the larger the distance value corresponding to the first different character pair is, and the larger the editing distance between the calculated character string and the word in the preset lexicon is. For example, if the relative position information of the first character and the second character on the keyboard is left-right adjacent, the distance value corresponding to the first different character pair is determined to be 0.1; if the relative position information of the first character and the second character on the keyboard is diagonally adjacent, determining that the distance value corresponding to the first different character pair is 0.4; and if the relative position information of the first character and the second character on the keyboard is not adjacent, determining that the distance value corresponding to the first different character pair is 1, thus determining the distance value corresponding to the first different character pair according to the relative position information, and further calculating the editing distance between the character string and the word in the preset lexicon according to the distance value.
For example, the character string input by the user is lock, the words in the preset lexicon are lock, comparing the characters in the lock with the characters at the corresponding positions in the words, determining that a first character "c" in the character string lock and a second character "a" in the words, and a third character "k" in the character string lock and a fourth character "m" in the words, form a second different character pair, and then determining that the relative position information of the first different character pair on the keyboard is not adjacent, and simultaneously determining that the relative position information of the second different character pair on the keyboard is diagonally adjacent, and according to the relative position information, respectively determining that the distance value corresponding to the first different character pair is 1 and the distance value corresponding to the second different character pair is 0.4, and finally adding the distance values corresponding to the first different character pair and the second different character pair to obtain the editing distance between the character string lock and the word "lock".
105. And displaying the words meeting the preset conditions as word suggestions based on the editing distance.
For the embodiment of the present invention, in order to show word suggestions to a user, an edit distance between a character string input by the user and each word in a preset lexicon needs to be calculated, then each word in the preset lexicon is sorted according to the size of the edit distance, words with sorting ranks within a preset ranking range (for example, the top N words) are screened according to the sorting result and recommended to the user as word suggestions, specifically, each word in the preset lexicon can be sorted from small to large according to the calculated edit distance, and words meeting preset conditions are used as word suggestions (for example, a word with an edit distance smaller than the preset edit distance is used as a word suggestion); the words in the preset word bank can be sorted in a reverse order from large to small according to the editing distance, and then the words (for example, the last N words) meeting the preset ranking condition are screened out as word suggestions according to the sorting result, or the words (for example, the words with the editing distance smaller than the preset editing distance) meeting the preset condition can be recommended to the user as word suggestions. Alternatively, part of the words in the obtained word suggestions may be recommended to the user, for example, when there are N (e.g., 10) words in the obtained word suggestions, the edit distance of m (e.g., 6) words is less than or equal to the target edit distance (the target edit distance is less than the preset edit distance), and thus the m words are recommended to the user as filtered word suggestions. It should be noted that, in the foregoing embodiment, in order to reduce the overall time taken to calculate the editing distance, the words may be sorted in the order from small to large according to the editing distance, so that when N required words with editing distances smaller than the preset editing distance are obtained through calculation, the continuous calculation of the editing distances between the character string and other words in the preset lexicon is stopped, the calculation of the editing distance corresponding to each word in the preset lexicon is avoided, and therefore, the calculation time is saved. When the number of words with editing distances smaller than the preset editing distance is large, the N words with the smaller editing distances can be taken as word suggestions.
In a specific application scene, if a character string input by a user is a log, words in a preset word bank are respectively a log and a load, an editing distance between the character string log and the word log is 0.4 through calculation, the editing distance between the character string log and the word load is 1, and the words are sorted from small to large according to the editing distance.
Further, a user can input a character string in a search box in the searching process, edit distances between the input character string and words in a preset word bank are calculated, the edit distances are sorted from small to large, words meeting preset conditions are screened according to sorting results and displayed in a display area corresponding to the search box, specifically, the display area and the display type of the corresponding words meeting the preset conditions can be selected according to the position information of the search box in a screen, wherein the display type can comprise a pull-up list and a pull-down list, and for example, if the search box is located in the middle upper portion of the display area of the screen, the words meeting the preset conditions can be displayed in the pull-down list; if the search box is in the middle-lower part of the screen display area, words meeting preset conditions can be displayed in the pull-up list. In addition, a mode that the pull-up list and the pull-down list exist at the same time can be adopted, and the words meeting the preset condition are alternately displayed in the pull-up list and the pull-down list according to the sequence of the words meeting the preset condition, for example, the word with the first sequence in the words meeting the preset condition can be displayed at the first position of the pull-down list, the word with the second sequence in the words meeting the preset condition can be displayed at the first position of the pull-up list, and the like.
In a specific application scenario, when a first different character pair appears, the voice information of the user can be acquired, and a word corresponding to the voice information is added to the word suggestion for display, based on which, the method further comprises: when the first different character pair is matched, prompting a user to perform voice input; determining words corresponding to the voice information input by the user, and adding the words corresponding to the voice information into the word suggestions for displaying, wherein the display sequence of the words corresponding to the voice information in the word suggestions is in a preset front display range. Specifically, when the first different matching pair appears, the loudspeaker icon is displayed to prompt the user to perform voice input, voice information input by the user is recognized, a word corresponding to the voice information is determined, the word corresponding to the voice information is added into the word suggestion to be displayed together, and for example, the word corresponding to the voice information is placed in the first three words in the word suggestion to be displayed.
Further, in the process of recommending the word suggestions to the user, pronunciations of words in the word suggestions can be played to the user in a voice playing mode, the words in the word suggestions can be displayed in a display area of a screen, and in order to enhance user experience, voice playing can be simultaneously performed on the user in the process of displaying the words in the display area of the screen.
For the embodiment of the present invention, after the edit distance between the character string and each word in the preset lexicon is calculated, in order to ensure the recommendation accuracy of the information, before the words are sorted according to the edit distance, the calculated edit distance also needs to be corrected, in a specific application scenario, the character string currently input by the user can be matched with the word prefix in the preset lexicon, whether the character string input by the user is a word prefix is judged, if the character string is a word prefix, the edit distance between the character string and each word containing the word prefix is corrected to 0, after the words are sorted according to the corrected edit distance, according to the sorting result, the words whose sorting ranking names are within the preset range are screened and recommended to the user, based on this, step 104 specifically includes: judging whether the character string is a word prefix in the preset word bank; if the character string is a word prefix, correcting the editing distance between the character string and each word containing the word prefix to be 0; and sequencing the words from small to large based on the corrected editing distance, and displaying the words meeting the preset conditions as word suggestions. Thus, by determining whether or not the character string is a word prefix, the calculated edit distance can be corrected, and the accuracy of word recommendation can be further improved.
Optionally, in the embodiment of the present invention, after the editing distance is corrected, if a word whose number exceeds a preset number threshold and whose corrected editing distance is 0 is obtained, the obtained word is screened as a word in the word suggestion according to the following manner: the words are sorted according to the word length corresponding to the word whose edit distance is 0 after the correction, for example, sorting the words with the editing distance of 0 after the correction according to the word length from short to long, screening the words with the editing distance of 0 after the correction of N before the ranking, or sorting the words with the editing distance of 0 after correction according to the word length from long to short, screening the words with the editing distance of 0 after correction of N after ranking, and/or, querying the historical query records of the user according to the words with the editing distance of 0 after correction, adding a preset number (for example, p) of words which are not in the historical query records of the user in the words with the editing distance of 0 after correction into the word suggestions, the words which are inquired by the user are recorded in the historical inquiry record, and the words which are not inquired by the user are added into the word suggestions and recommended to the user together because the user is unfamiliar with the words which are not inquired. Meanwhile, a word with the next character position order before the corrected edit distance 0 can be screened out from the words with the corrected edit distance 0 according to the position order of the next character corresponding to the character string input by the user in the alphabet, for example, the character string input by the user is lo, the sequence of the next character in the alphabet is a, c, g, n, o, s, v, w, and a word with the next character position order before the corrected edit distance 0 in the alphabet is screened out according to the position order of the next character after the corrected edit distance 0 in the alphabet, if the word before the screened position order is load, local, log, long, and hook, the screened word is added to the word suggestion and recommended to the user, and the word with the next character in the alphabet after the position order is ignored.
It should be noted that, the editing distance may be corrected after the editing distance between the character string and each word is calculated, and after the character string input by the user is obtained, it may be directly determined whether the character string is a word prefix, and if the character string is a word prefix, the editing distance between the character string and each word containing the word prefix in the preset lexicon is directly determined to be 0; if the character string is not the word prefix, the relative position information of the characters in the character string and the corresponding characters in the words on the character input device is continuously determined, and the editing distance between the character string and the words is calculated according to the relative position information, so that the editing distance can be directly determined and the calculation process is reduced under the condition that the character string is the word prefix.
Optionally, in the embodiment of the present invention, after obtaining the character string input by the user, directly determining that the character string is a word prefix, and when obtaining words with an editing distance of 0 whose number exceeds a preset number threshold, the word obtained by screening according to the following manner is used as a word in a word suggestion: the words are sorted according to the word length corresponding to the word with the editing distance of 0, for example, the words with the editing distance of 0 are sorted according to the word length from short to long, the words with the editing distance of 0 before ranking are screened, or the words with the editing distance of 0 are sorted according to the word length from long to short, the words with the editing distance of 0 after ranking are screened, and/or the historical query records of the user are queried according to the words with the editing distance of 0, and the words which are not in the historical query records of the user and exist in the words with the editing distance of 0 are added into word suggestions, wherein the words which are queried by the user are recorded in the historical query records, and the words which are not queried by the user are added into the word suggestions and recommended to the user together because the user is stranger about the words which are not queried. Meanwhile, a word with the position order of the next character in the alphabet, which is the first character position order, can be screened from words with the editing distance of 0 according to the position order of the next character corresponding to the character string input by the user in the alphabet, for example, the character string input by the user is lo, the sequence of the next character in the alphabet is a, c, g, n, o, s, v, w, and a word with the position order of the next character in the alphabet, which is the editing distance of 0, is located after lo, is screened from words with the position order of the next character in the preceding M in the editing distance of 0, and if the word in the preceding M in the selected position order is local, log, long, and hook, the screened words are added to the word suggestion and recommended to the user, while the word with the position order of the next character in the alphabet is ignored.
When words meeting preset conditions are displayed as word suggestions, for words with the same editing distance (including but not limited to words with the editing distance of 0 when the character string is a word prefix), the display priority of the words with the same editing distance can be determined as follows: the words with the same editing distance are sorted according to the word lengths corresponding to the words with the same editing distance, the display priority corresponding to the words with the same editing distance is determined according to the sorting ranking, specifically, the words with the same editing distance can be sorted from short to long according to the word lengths corresponding to the words with the same editing distance, the display priority corresponding to the word with the higher ranking is higher, or the words with the same editing distance can be sorted from long to short according to the word lengths corresponding to the words with the same editing distance, and the display priority corresponding to the word with the higher ranking is higher. Meanwhile, the display priority corresponding to the word with the same edit distance may be determined according to the position order of the next character in the word with the same edit distance in the alphabet, where the next character is the next character in the word immediately after the character string, if the character string input by the user is lo, the next character in the word load immediately after the character string is "a", the next character in the word local immediately after the character string is "c", the next character in the word log immediately after the character string is "g", the position order of the next character in the character table is a, c, g by referring to the alphabet, and the display priority corresponding to the word load is determined to be the highest, the next character in the word log is determined to be the word local, the display priority corresponding to the word log is the lowest, that is, the word string before the position order of the next character in the word after the input character string is the higher, the words with the higher display priority are displayed before the words with the lower display priority, or the display position corresponding to the word with the higher display priority is positioned before the display position corresponding to the word with the lower display priority.
In a specific application scenario, the calculated edit distance may be further modified according to the current knowledge level of the user, and based on this, the word meeting the preset condition is displayed based on the edit distance, including: acquiring the current knowledge level of a user; determining words matched with the current knowledge level in the words, and downwards correcting the editing distance corresponding to the words matched with the current knowledge level; and sequencing the words from small to large based on the corrected editing distance, and displaying the words meeting the preset conditions. Wherein, the current knowledge level of the user comprises English level four, English level six, Yasi, Toufu and the like. For example, the current knowledge level of the user is level four english, level four english words in a preset word bank are determined, the editing distance between the character string and the level four english words is corrected downwards, if the editing distance is reduced by 0.5, further, the words are sorted according to the corrected editing distance, and according to a sorting result, the words with the sorting ranking in the preset sorting range are screened and recommended to the user. Optionally, words with the modified edit distance smaller than the preset edit distance may be screened out and added to the word suggestions to be recommended to the user, further, in order to ensure that the recommended words can better meet the search requirements of the user, words in the word suggestions may be further screened, and optionally, a part of words in the word suggestions may be recommended to the user, for example, 10 words are in the current word suggestions, wherein the edit distance of 6 words is smaller than or equal to the target edit distance (wherein the target edit distance is smaller than the preset edit distance), so that the 6 words are recommended to the user as the screened words. Therefore, the calculated editing distance can be corrected according to the current knowledge level of the user, so that the recommended word information can be ensured to conform to the current knowledge level of the user, and the recommended word information can be ensured to meet the search requirement of the user.
In a specific application scenario, the calculated edit distance may also be adjusted according to a historical search record of a user, and based on this, displaying words meeting a preset condition based on the edit distance includes: acquiring a historical search record of a user; and correcting the editing distance according to the historical search record, sequencing the words from small to large based on the corrected editing distance, and displaying the words meeting preset conditions. The historical search records record the corresponding relation between the character strings input by the user and the click words in the word list of the search suggestions.
For example, the character string input by the user is loak, the presentation position of the word loam recommended in the word suggestion is better than the presentation position of the word lock, since the edit distance between the word "loam" and the character string "leak" is smaller than the edit distance between the word "lock" and the character string "leak" as calculated by the edit distance, but through counting the historical search records of the user, when the user inputs 'lock' for a plurality of times, the word 'lock' is clicked by the user, the edit distance between the character string "lock" and the word "lock" is adjusted downward to ensure that the edit distance between the character string "lock" and the word "lock" is less than the edit distance between the character string "lock" and the word "lock", the word "lock" is put to the better position and is shown in order to make things convenient for the user to click, can guarantee that the word information of recommending simultaneously satisfies user's search demand.
Compared with the mode that once the input characters are inconsistent with the corresponding characters in the words at present, no association exists between the two characters, the method for displaying the word suggestions provided by the embodiment of the invention can obtain the character strings input by the user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; at the same time, determining relative position information of the first different character pair on the character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; finally, words meeting preset conditions are displayed as word suggestions based on the editing distance, the editing distance between the character string and the words is calculated according to the relative position information of the first different character pairs on the character input keyboard, the layout factors of the character input keyboard can be referred in the calculation process of the editing distance, the calculation precision of the editing distance between the character string and the words is improved, the words meeting the preset conditions are recommended to a user as the word suggestions according to the editing distance, the word recommendation precision in the word searching process can be improved, the word suggestions can meet the searching requirements of the user, and the user experience is enhanced.
Further, in order to better explain the search process of the word, as a refinement and extension to the above embodiment, another method for presenting word suggestions is provided in an embodiment of the present invention, as shown in fig. 2, where the method includes:
201. and acquiring the character string input by the user.
For the embodiment of the invention, the user can search words in various English-related APPs, and the user can input the corresponding English character string in the search box during specific search, so that the character string input by the user can be acquired, and the character string can be matched with each word in the preset word bank.
202. And matching the first character in the character string with the second character in each word in a preset word bank to obtain a first different character pair.
For the embodiment of the present invention, in order to obtain the first different character pair, step 202 specifically includes: judging whether the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word in each word: if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, matching a first character in the character string with a second character in the target word to obtain a first different character pair; and if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, inserting or deleting the target characters in the character string, and matching a first character in the character string after the target characters are inserted or deleted with a second character in the target word to obtain a first different character pair. The target word can be any word in a preset word bank.
For example, the character string currently input by the user is loak, the target word in the preset lexicon is loam, and since the number of characters corresponding to the character string loak is the same as the number of characters corresponding to the target word loam, the first character "k" in the loak can be directly matched with the second character "m" in the target word loam to obtain a first character pair, and further, the relative position information of the first character "k" and the second character "m" on the keyboard is determined to be diagonally adjacent by querying a preset keyboard character position information table.
For another example, the character string currently input by the user is lukt, the target word in the preset lexicon is lucky, since the number of characters corresponding to the character string lukt is different from the number of characters corresponding to the destination word lucky, therefore, the character matching cannot be directly performed, the target character "c" needs to be inserted into the character string lukt first, so that the number of characters corresponding to the character string after the target character is inserted is the same as the number of characters corresponding to the target word lucky, and simultaneously the character string after the target character is inserted is ensured to be closest to the target word lucky, thus determining that the character string after the target character is inserted is luckt, further, matching a first character "t" in the character string after the target character is inserted with a second character "y" in the lucky to obtain a first different character pair, and determining that the relative position information of the first character "t" and the second character "y" of the target word on the keyboard is left-right adjacent by inquiring a preset keyboard character position information table. Therefore, the first different character pair can be obtained under different scenes according to the mode, and then the relative position information of the first different character pair on the character input keyboard can be determined, so that the editing distance between the character string and the target word can be calculated according to the relative position information.
Further, for a specific process of inserting or deleting a target character in a character string, as an optional implementation, the method includes: comparing the character string with the target word, and determining the target character inserted or deleted in the character string according to a comparison result; and inserting or deleting the target character in the character string, so that the similarity between the character string after the target character is inserted or deleted and the target word is larger than a preset threshold value. The preset threshold value can be set according to the service requirement of word search, and it should be noted that, in order to ensure the precision of word search, the preset threshold value is not suitable to be set too low.
For example, a character string input by a user is beautiful, a target word to be matched in a preset lexicon is beautiful, the number of characters corresponding to the character string is 8, and the number of characters corresponding to the target word is 9, so that it is determined that the number of characters corresponding to the character string is less than the number of characters corresponding to the target word, 1 corresponding character needs to be inserted into the character string, specifically, in order to ensure that the character string after inserting the character is closest to the target word, that is, the similarity between the character string after inserting the character and the target word is greater than a preset threshold, a target character "t" is inserted between a character "u" and a character "i" of the character string, and the character string after inserting the target character is completely the same as the target word.
203. Determining relative position information of the first different character pair on a character entry keyboard.
For the embodiment of the present invention, in order to calculate the edit distance between the character string and each word, the relative position information of the first different character pair on the character input keyboard needs to be determined, and the specific process for determining the relative position information is completely the same as that in step 103, and is not described herein again.
204. And calculating the editing distance between the character string and each word according to the relative position information.
For the embodiment of the present invention, if the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word, the edit distance between the character string and the target word may be directly calculated according to the distance value of the first different character pair, based on which step 204 specifically includes: if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, determining a distance value corresponding to the first different character pair according to the relative position information; and determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair. The calculation process of the edit distance in this scenario is completely the same as the calculation process of the edit distance in step 104, and is not described herein again.
Further, if the number of characters corresponding to the character string is different from the number of characters corresponding to the target word, when calculating the edit distance, not only the distance value corresponding to the first different character pair but also an operation of inserting or deleting a character in the character string need to be considered, based on which step 204 specifically includes: if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, determining a first editing distance corresponding to the character string according to the number of the characters corresponding to the target character; determining a second editing distance corresponding to the character string according to the relative position information; and adding the first editing distance and the second editing distance to obtain the editing distance between the character string and the target word.
For example, the character string input by the user is beeutfill, the target word to be matched in the preset lexicon is beautifull, since the number of characters corresponding to the character string is less than the number of characters corresponding to the target word, it is necessary to insert characters at corresponding positions in the character string, specifically, one character "i" is inserted between the character "t" and the character "f" of the character string, the character string after inserting the character is beeutfull, since only one character is inserted in the character string, it can be determined that the number of characters corresponding to the character string after inserting the character is the same as the number of characters corresponding to the target word through one insertion operation, and further, according to the number of the inserted characters, the first editing distance corresponding to the character string is determined to be 1, further, the first character "e" in the character string after inserting the character is matched with the second character "a" in the target word, obtaining a first different character pair, matching a third character "i" in the character string after inserting the characters with a fourth character "u" in the target word to obtain a second different character pair in the same way, determining that the relative position information of the first character "e" and the second character "a" in the target word on the keyboard is not adjacent by inquiring a preset keyboard character position information table, simultaneously determining that the relative position information of the third character "i" and the fourth character "u" in the target word on the keyboard is left-right adjacent, thereby determining that the distance value corresponding to the first different character pair is 1 and the distance value corresponding to the second different character pair is 0.1 according to the relative position information, further adding the distance values corresponding to the first different character pair and the second different character pair to obtain a second editing distance corresponding to the character string is 1.1, and finally adding the calculated first editing distance and the second editing distance, the editing distance between the character string beeutfill and the target word beautifull is obtained to be 2.1, so that when the number of characters corresponding to the character string is different from the number of characters corresponding to the target word, the editing distance between the character string and the target word can be calculated according to the method, the influence caused by keyboard layout factors is fully considered in the process of calculating the editing distance, and the calculation accuracy of the editing distance is improved.
In a specific application scenario, as a user inputs characters one by one, according to feedback of recommended content, the user may continuously correct characters that are input before, so that a probability of making an error in a character that is positioned earlier in a currently obtained character string is lower, and a probability of making an error in a character that is positioned later is higher, and thus, in a process of calculating an edit distance between a character string and a target word, not only a factor of a keyboard layout needs to be considered, but also a correct rate of inputting characters at different positions in the character string needs to be considered, based on which, after a distance value corresponding to the first different character pair is obtained according to the relative position information, the method further includes: determining the correct rate weight corresponding to the first different character pair according to the position information of the first character in the character string; meanwhile, the determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair includes: determining an edit distance between the character string and the target word based on the correctness weight and the distance value.
Specifically, because the probability that the character input at the position closer to the front in the character string is correct is higher, the accuracy rate weight corresponding to the character can be determined according to the position information corresponding to the character in the character string, where the accuracy rate weight represents the possibility that the character user at the position inputs the character correctly, and the accuracy rate weight is higher, the possibility that the character input is correct is higher, specifically, the position information corresponding to the character can be input into a preset accuracy rate weight prediction model for weight prediction, so as to obtain the accuracy rate weight corresponding to the character, and the preset accuracy rate weight prediction model can be: and f (s, i) — (1.0+ s.length-i)/(4.0+ s.length), where s.length is the length of the character string, i.e. the number of characters included in the character string, and i is an index of a subscript position of the character in the character string, i.e. position information corresponding to the character, such as 0, 1, 2, etc., so that the correct rate weight corresponding to the character can be determined according to the position information corresponding to the character and a preset correct rate weight prediction model.
For example, the character string input by the user is lock, the target word in the preset lexicon is lock, the first different character pair "c" and "a" and the second different character pair "k" and "m" are determined by comparing the character string with the target word, the distance value corresponding to the first different character pair is 1 and the distance value corresponding to the second different character pair is 0.4, further, the correct rate weight corresponding to the first character "c" is 3/8 according to the position information 3 corresponding to the first character "c" and the preset correct rate weight prediction model, the correct rate weight corresponding to the third character "k" is 2/8, further, in the process of calculating the edit distance between the character string and the target word, the correct rate weight corresponding to the first character "c" and the distance value are multiplied by considering the keyboard layout factor and the factor of the correct rate input by the user, the multiplication result corresponding to the first character "c" is obtained as 3/8, the correct rate weight corresponding to the third character "k" is multiplied by the distance value in the same way, the multiplication result corresponding to the third character "k" is obtained as 0.1, finally, the multiplication result corresponding to the first character "c" and the multiplication result corresponding to the third character "k" are added, the editing distance between the character string lock and the target word lock is obtained, and the keyboard layout factor and the user input correct rate factor can be considered at the same time according to the calculated editing distance, so that the calculation accuracy of the editing distance is further improved, and the reliability of the word suggestion is ensured.
In a specific application scenario, when the edit distance is calculated, historical error weights of first different character pairs may also be considered, for example, the first different character pairs are a and s, and the lower the historical error weight of the first different character pair is, the more historical error times indicating that the user mistakenly inputs the character a as the character s, the more the user increases the error frequency of the character each time the user increases, the historical error weight corresponding to the first different character pair may decrease, and based on this, after the distance value corresponding to the first different character pair is obtained according to the relative position information, the method further includes: inquiring a preset historical error weight table according to the first different character pair, and determining the historical error weight corresponding to the first different character pair; meanwhile, the determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair includes: determining an edit distance between the character string and the target word based on the historical error weight and the distance value. The preset historical error weight table records historical error weights corresponding to different character pairs.
For example, when the character string input by the user at this time is "loak", and the edit distance between the character string and the word "look" is calculated, firstly, a preset historical error weight table is queried, the historical error weights corresponding to the first different character pair "a" and "o" are determined, then, the historical error weight corresponding to the first different character pair is multiplied by the distance value, so as to obtain the edit distance between the character string "look" and the word "look", and when the number of times that the user mistakes the first character "a" into the second character "o" is greater, the historical error weight corresponding to the first different character pair is smaller, and the edit distance calculated by the method is smaller. Therefore, the influence caused by the historical error weight can be considered in the process of calculating the edit distance, and the calculation precision of the edit distance is further improved.
Further, in the process of calculating the edit distance, the influence of the correct rate weight and the historical error weight can be considered at the same time, specifically, the correct rate weight, the historical error weight and the distance value corresponding to a first different character pair can be multiplied to obtain the edit distance between the character input by the user and the word, if a second different character pair exists, the correct rate weight, the historical error weight and the distance value corresponding to the second different character pair are multiplied, and the multiplication result corresponding to the second different character pair is added to the multiplication result corresponding to the first different character pair to obtain the edit distance between the input character and the word.
205. And taking the words with the editing distance smaller than the preset editing distance in the words as word suggestions.
The preset editing distance can be set according to actual business requirements, for example, words with the editing distance smaller than 1.2 are added into word suggestions and displayed to a user, it needs to be stated that the preset editing distance is not suitable to be set too large in order to guarantee recommendation precision.
206. And determining the display priority corresponding to the words in the word suggestions according to the editing distance corresponding to the words in the word suggestions.
For the embodiment of the invention, the smaller the editing distance corresponding to the word in the word suggestion is, the higher the display priority corresponding to the word is, and conversely, the larger the editing distance corresponding to the word is, the lower the display priority corresponding to the word is, for example, 10 words exist in the word suggestion, the display priority corresponding to the word in the word suggestion is determined according to the sequencing result after the 10 words are sequenced from small to large according to the editing distance corresponding to the 10 words.
207. And displaying the words in the word suggestions based on the display priority.
Wherein the presentation priority is used to indicate a presentation position and/or a presentation order of the words in the word suggestion. For example, in presenting word suggestions, words with a high presentation priority are presented before words with a low presentation priority and/or the presentation position of words with a high presentation priority is located before words with a low presentation priority.
For the embodiment of the present invention, in the process of displaying the words in the word suggestions, the display priority of the words may be adjusted according to the current learning task of the user, and based on this, step 207 specifically includes: determining words matched with the user learning task in the word suggestions, and correcting the display priority corresponding to the words matched with the user learning task; and displaying the words in the word suggestions based on the corrected display priority.
For example, the current learning task of the user is the learning of a fourth-level word, the display priority of the fourth-level word in the word suggestion is improved, specifically, the display position of the fourth-level word can be advanced in the display area of the screen, or the pronunciation corresponding to the fourth-level word is preferentially played, in addition, the fourth-level word can be also played in a voice mode while being preferentially displayed in the display area of the screen, and the user experience is enhanced. For another example, if the current learning task of the user is learning of words outside the four-level range, the display priority corresponding to the words outside the four-level range in the word suggestion can be increased, so that the words outside the four-level range are preferentially recommended to the user, and the user can conveniently learn other words.
In a specific application scenario, the display priority of the word in the word suggestion may also be modified according to the historical click record of the user, and based on this, step 207 specifically includes: if the user history clicking record corresponding to the character string exists, determining words existing in the user history clicking record in the word suggestion; according to the click frequency and/or click time corresponding to the words existing in the user history click record in the word suggestions, correcting the display priority corresponding to the words existing in the user history click record in the word suggestions; and displaying the words in the word suggestions based on the corrected display priority. When a character string input by a user is recorded in the history clicking record of the user, a word selected from the word suggestions is a clicked word, and the clicking frequency (i.e., the number of times the user clicks the word) and/or the clicking time (i.e., the time the user clicks the word) of each clicked word are/is determined.
For example, the user enters a string of loaks, the words "look" and "load" in the word suggestions exist in the user's historical click record, determining the clicking frequency and/or the clicking time respectively corresponding to the words "look" and "load" by inquiring the historical clicking records of the user, in the process of displaying the words, if the click frequency corresponding to the word "look" is higher than the click frequency corresponding to the word "load", the presentation priority corresponding to the word "load" is modified to be higher than the presentation priority corresponding to the word "look" (i.e. the presentation priority of words with higher frequency of clicks is lower), alternatively, if the click time corresponding to the word "look" is later than the click time corresponding to the word "load", the presentation priority corresponding to the word "load" is modified to be higher than the presentation priority corresponding to the word "look", i.e., words that the user has queried at an earlier time have a higher presentation priority than words that have been queried at a later time.
Further, if the user has not input the character string before, there is no history click record of the user for the character string, and at this time, the display priority of the word may be modified according to the history click record of the other user for the character string, based on which step 207 specifically includes: if the user history clicking record corresponding to the character string does not exist, collecting history clicking records of other users, and sequencing the clicked words according to clicking frequencies corresponding to the clicked words in the history clicking records of the other users; determining target point word selection with the ranking rank within a preset range according to the ranking result; judging whether the target point word selection exists in the word suggestions or not; if the target point word selection exists in the word suggestions, correcting the display priority corresponding to the target point word selection; and displaying the words in the word suggestions based on the corrected display priority. Further, after the determining whether the target point vote exists in the word suggestions, the method further includes: and if the target point vote does not exist in the word suggestions, adding the target point vote meeting the preset top ranking condition into the word suggestions, and displaying the words in the added word suggestions. The collected historical click records of other users are click records of other users when inputting the character string, that is, click words and click times of other users when inputting the character string, and click frequencies corresponding to the click words are recorded.
For example, the user enters the string "loak" for the first time, since there is no historical click record for the string by the user, therefore, historical click records of other users aiming at the character string "loak" are collected, click frequencies corresponding to click words in the historical click records of the other users are counted, sorting the word hits according to the frequency of word hits, screening the word hits n before the ranking according to the sorting result, further, determining whether there is a word hit in the word suggestions, if the word suggestion has the target point vote, the display priority corresponding to the target point vote in the word suggestion is corrected, if m click vocabularies before the ranking are in the word suggestion, the display priority corresponding to the m point votes is increased, and the higher the point vote frequency of the m point votes is, the higher the display priority corresponding to the point votes is, the lower the point vote frequency is, the lower the display priority corresponding to the point votes is; if no target point vote exists in the word suggestions, namely n click words before the ranking do not exist in the word suggestions, sorting is carried out according to the sequence of the click frequency from large to small, click words which accord with the condition that the preset ranking is near the top are screened from the n click words and added into the word suggestions for displaying, for example, 3 click words before the ranking are screened from the n click words and added into the word suggestions for displaying, and the display priority of the click words is determined according to the click frequency corresponding to the click words during specific displaying.
In a specific application scenario, during the process of learning words, a user may mark learned state information for the learned words, such as a word being marked as mastered, a learned solution, strange, and the like, and then determine a corresponding display priority according to the learning state information corresponding to each word in the word suggestion, where the word with a worse user mastered degree has a higher corresponding display priority, the word with a better user mastered degree has a lower corresponding display priority, and based on this, step 207 specifically includes: acquiring learning state information of word marks in the word suggestions; based on the learning state information, correcting the display priority corresponding to the words in the word suggestions; and displaying the words in the word suggestions based on the corrected display priority.
For example, the word input by the user is hook, the words marked in the word suggestions are hook and load, the learning state information corresponding to the word hook is a known solution, and the learning state information corresponding to the word load is unknown, so that the presentation priority corresponding to the word load is adjusted to be higher than the presentation priority corresponding to the word hook.
Compared with the mode that once the input characters are inconsistent with the corresponding characters in the words at present, no association exists between the two characters, the method for displaying the word suggestions provided by the embodiment of the invention can obtain the character strings input by the user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; at the same time, determining relative position information of the first different character pair on the character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; finally, words meeting preset conditions are displayed as word suggestions based on the editing distance, the editing distance between the character string and the words is calculated according to the relative position information of the first different character pairs on the character input keyboard, the layout factors of the character input keyboard can be referred in the calculation process of the editing distance, the calculation precision of the editing distance between the character string and the words is improved, the words meeting the preset conditions are recommended to a user as the word suggestions according to the editing distance, the word recommendation precision in the word searching process can be improved, the word suggestions can meet the searching requirements of the user, and the user experience is enhanced.
Further, as a specific implementation of fig. 1, an embodiment of the present invention provides an apparatus for presenting word suggestions, as shown in fig. 4, the apparatus includes: an acquisition unit 31, a matching unit 32, a determination unit 33, a calculation unit 34 and a presentation unit 35.
The acquiring unit 31 may be configured to acquire a character string input by a user.
The matching unit 32 may be configured to match a first character in the character string with a second character in each word in a preset lexicon to obtain a first different character pair.
The determining unit 33 may be configured to determine relative position information of the first different character pair on the character input keyboard.
The calculating unit 34 may be configured to calculate an edit distance between the character string and each word according to the relative position information.
The display unit 35 may be configured to display, as word suggestions, words that meet preset conditions based on the edit distance.
In a specific application scenario, in order to obtain a first different character pair, as shown in fig. 5, the matching unit 32 includes: a decision module 321, a matching module 322 and an insertion module 323.
The determining module 321 may be configured to determine whether the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word in the words.
The matching module 322 may be configured to match a first character in the character string with a second character in the target word to obtain a first different character pair if the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word.
The inserting module 323 may be configured to insert or delete a target character in the character string if the number of characters corresponding to the character string is different from the number of characters corresponding to the target word, and match a first character in the character string after the target character is inserted or deleted with a second character in the target word to obtain a first different character pair.
Further, in order to determine a target character to be inserted or deleted in the character string, the insertion module 323 includes: a comparison submodule and an insertion submodule.
The comparison submodule may be configured to compare the character string with the target word, and determine the target character to be inserted or deleted in the character string according to a comparison result.
The inserting sub-module may be configured to insert or delete the target character in the character string, so that a similarity between the character string after the target character is inserted or deleted and the target word is greater than a preset threshold.
In a specific application scenario, in order to calculate an edit distance between the character string and the target word, the calculating unit 34 includes: a first determination module 341 and a second determination module 342.
The first determining module 341 may be configured to determine, according to the relative position information, a distance value corresponding to the first different character pair if the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word.
The second determining module 342 may be configured to determine an edit distance between the character string and the target word according to the distance value corresponding to the first different character pair.
In a specific application scenario, in order to calculate an edit distance between the character string and the target word, the calculating unit 34 includes: a first determination module 341, a second determination module 342, and an addition module 343.
The first determining module 341 may be configured to determine, if the number of characters corresponding to the character string is different from the number of characters corresponding to the target word, a first edit distance corresponding to the character string according to the number of characters corresponding to the target character.
The second determining module 342 may be configured to determine a second edit distance corresponding to the character string according to the relative position information.
The adding module 343 may be configured to add the first editing distance and the second editing distance to obtain an editing distance between the character string and the target word.
Further, in order to determine an editing distance between a character string and a target word, the first determining module may be further configured to determine, according to position information of the first character in the character string, a correct rate weight corresponding to the first different character pair; and/or querying a preset historical error weight table according to the first different character pair, and determining the historical error weight corresponding to the first different character pair.
The second determining module may be specifically configured to determine an edit distance between the character string and the target word based on the correctness weight and/or the historical error weight, and the distance value.
In a specific application scenario, in order to display words in the word suggestions, the display unit 35 includes: a judgment module 351, a correction module 352 and a presentation module 353.
The determining module 351 may be configured to determine whether the character string is a word prefix in the preset lexicon.
The correcting module 352 may be configured to correct the edit distance between the character string and each word including the word prefix to 0 if the character string is the word prefix.
The display module 353 may be configured to sort the words from small to large based on the modified edit distance, and display the words meeting the preset condition as word suggestions.
In a specific application scenario, in order to perform voice playing for a user, the apparatus further includes: a presentation unit 36.
The prompting unit 36 may be configured to prompt the user to perform voice input when the first different character pair is matched.
The display unit 35 may be further configured to determine a word corresponding to the voice information input by the user, and add the word corresponding to the voice information into the word suggestion to be displayed, where a display sequence of the word corresponding to the voice information in the word suggestion is in a preset front display range.
In a specific application scenario, in order to display words in the word suggestions, the display unit 35 includes: a determination module 354 and a presentation module 353.
The determining module 354 may be configured to use, as a word suggestion, a word of which an editing distance is smaller than a preset editing distance in the words;
the determining module 354 may be further configured to determine, according to the editing distance corresponding to the word in the word suggestion, a display priority corresponding to the word in the word suggestion;
the presentation module 353 may be configured to present the words in the word suggestions based on the presentation priority.
In a specific application scenario, the display priority of the word may be modified based on the current learning task of the user, and the display module 353 includes: a correction submodule and a display submodule.
The modification submodule may be configured to determine a word in the word suggestion that matches the user learning task, and modify a display priority corresponding to the word that matches the user learning task.
The display submodule may be configured to display the word in the word suggestion based on the modified display priority.
In a specific application scenario, the display priority of the word may be modified based on the historical click record of the user, and the display module 353 includes: the device comprises a determination submodule, a correction submodule and a display submodule.
The determining sub-module may be configured to determine, if a user history click record corresponding to the character string exists, a word in the user history click record in the word suggestion.
The modification sub-module may be configured to modify, according to the click frequency and/or the click time corresponding to a word existing in the user history click record in the word suggestion, a presentation priority corresponding to a word existing in the user history click record in the word suggestion.
The display submodule may be configured to display the word in the word suggestion based on the modified display priority.
In a specific application scenario, the display priority of the word may also be modified based on historical click records of other users, where the display module 353 includes: the device comprises a sequencing submodule, a determining submodule, a judging submodule, a correcting submodule and a displaying submodule.
The sorting submodule is used for collecting historical click records of other users if the user historical click record corresponding to the character string does not exist, and sorting the click words according to the click frequency corresponding to the click words in the historical click records of the other users;
the determining sub-module may be configured to determine, according to the sorting result, a target point whose sorting rank is within a preset range for word selection.
The determining sub-module may be configured to determine whether the target point vote exists in the word suggestion.
The modification sub-module may be configured to modify the display priority corresponding to the target word choice if the target word choice exists in the word suggestion.
The display submodule may be configured to display the word in the word suggestion based on the modified display priority.
Further, the display module 353 further includes: and the adding sub-module is used for adding the target point vote meeting the preset top ranking condition into the word suggestion if the target point vote does not exist in the word suggestion, and displaying the word in the added word suggestion.
In a specific application scenario, the display priority of the word may be modified based on the learning state information of the word label, and the display module 353 includes: the system comprises an acquisition submodule, a correction submodule and a display submodule.
The obtaining sub-module may be configured to obtain learning state information of word tokens in the word suggestion.
The correction submodule may be configured to correct a display priority corresponding to a word in the word suggestion based on the learning state information.
The display submodule may be configured to display the word in the word suggestion based on the modified display priority.
It should be noted that other corresponding descriptions of the functional modules related to the apparatus for presenting word suggestions provided in the embodiment of the present invention may refer to the corresponding descriptions of the method shown in fig. 1, and are not described herein again.
Based on the method shown in fig. 1, correspondingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the following steps: acquiring a character string input by a user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; determining relative position information of the first different character pair on a character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; and displaying the words meeting the preset conditions as word suggestions based on the editing distance.
Based on the above embodiments of the method shown in fig. 1 and the apparatus shown in fig. 4, an embodiment of the present invention further provides an entity structure diagram of a computer device, as shown in fig. 6, where the computer device includes: a processor 41, a memory 42, and a computer program stored on the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43 such that when the processor 41 executes the program, the following steps are performed: acquiring a character string input by a user; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; determining relative position information of the first different character pair on a character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; and displaying the words meeting the preset conditions as word suggestions based on the editing distance.
By the technical scheme, the character string input by the user can be acquired; matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair; at the same time, determining relative position information of the first different character pair on the character input keyboard; calculating the editing distance between the character string and each word according to the relative position information; finally, words meeting preset conditions are displayed as word suggestions based on the editing distance, the editing distance between the character string and the words is calculated according to the relative position information of the first different character pairs on the character input keyboard, the layout factors of the character input keyboard can be referred in the calculation process of the editing distance, the calculation precision of the editing distance between the character string and the words is improved, the words meeting the preset conditions are recommended to a user as the word suggestions according to the editing distance, the word recommendation precision in the word searching process can be improved, the word suggestions can meet the searching requirements of the user, and the user experience is enhanced.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method of presenting word suggestions, comprising:
acquiring a character string input by a user;
matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair;
determining relative position information of the first different character pair on a character input keyboard;
calculating the editing distance between the character string and each word according to the relative position information;
and displaying the words meeting the preset conditions as word suggestions based on the editing distance.
2. The method of claim 1, wherein matching a first character in the string of characters to a second character in each word in a predetermined lexicon to obtain a first different character pair comprises:
judging whether the number of characters corresponding to the character string is the same as the number of characters corresponding to the target word in each word:
if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, matching a first character in the character string with a second character in the target word to obtain a first different character pair;
and if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, inserting or deleting the target characters in the character string, and matching a first character in the character string after the target characters are inserted or deleted with a second character in the target word to obtain a first different character pair.
3. The method of claim 2, wherein the inserting or deleting a target character in the character string comprises:
comparing the character string with the target word, and determining the target character inserted or deleted in the character string according to a comparison result;
and inserting or deleting the target character in the character string, so that the similarity between the character string after the target character is inserted or deleted and the target word is larger than a preset threshold value.
4. The method of claim 2, wherein said calculating an edit distance between the character string and the words according to the relative position information comprises:
if the number of the characters corresponding to the character string is the same as the number of the characters corresponding to the target word, determining a distance value corresponding to the first different character pair according to the relative position information;
and determining the editing distance between the character string and the target word according to the distance value corresponding to the first different character pair.
5. The method of claim 2, wherein said calculating an edit distance between the character string and the words according to the relative position information comprises:
if the number of the characters corresponding to the character string is different from the number of the characters corresponding to the target word, determining a first editing distance corresponding to the character string according to the number of the characters corresponding to the target character;
determining a second editing distance corresponding to the character string according to the relative position information;
and adding the first editing distance and the second editing distance to obtain the editing distance between the character string and the target word.
6. The method of claim 4, wherein after obtaining the distance value corresponding to the first different character pair according to the relative position information, the method further comprises:
determining the correct rate weight corresponding to the first different character pair according to the position information of the first character in the character string; and/or
Inquiring a preset historical error weight table according to the first different character pair, and determining the historical error weight corresponding to the first different character pair;
determining an editing distance between the character string and the target word according to the distance value corresponding to the first different character pair, including:
determining an edit distance between the character string and the target word based on the correct rate weight and/or the historical error weight, and the distance value.
7. The method of claim 1, wherein presenting words meeting preset conditions as word suggestions based on the edit distance comprises:
judging whether the character string is a word prefix in the preset word bank;
if the character string is a word prefix, correcting the editing distance between the character string and each word containing the word prefix to be 0;
and sequencing the words from small to large based on the corrected editing distance, and displaying the words meeting the preset conditions as word suggestions.
8. An apparatus for presenting word suggestions, comprising:
the device comprises an acquisition unit, a display unit and a control unit, wherein the acquisition unit is used for acquiring a character string input by a user;
the matching unit is used for matching a first character in the character string with a second character in each word in a preset word bank to obtain a first different character pair;
a determining unit for determining relative position information of the first different character pair on a character input keyboard;
a calculation unit configured to calculate an edit distance between the character string and each word according to the relative position information;
and the display unit is used for displaying the words meeting the preset conditions as word suggestions based on the editing distance.
9. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 7 when executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
CN202110343650.6A 2021-03-30 2021-03-30 Method, device, storage medium and computer equipment for showing word suggestion Active CN112905026B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110343650.6A CN112905026B (en) 2021-03-30 2021-03-30 Method, device, storage medium and computer equipment for showing word suggestion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110343650.6A CN112905026B (en) 2021-03-30 2021-03-30 Method, device, storage medium and computer equipment for showing word suggestion

Publications (2)

Publication Number Publication Date
CN112905026A true CN112905026A (en) 2021-06-04
CN112905026B CN112905026B (en) 2024-04-16

Family

ID=76109574

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110343650.6A Active CN112905026B (en) 2021-03-30 2021-03-30 Method, device, storage medium and computer equipment for showing word suggestion

Country Status (1)

Country Link
CN (1) CN112905026B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113743102A (en) * 2021-08-18 2021-12-03 百度在线网络技术（北京）有限公司 Method and device for recognizing characters and electronic equipment

Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05175256A (en) * 1991-12-20 1993-07-13 Hitachi Ltd Pellet bonding device
US20060004744A1 (en) * 2004-06-19 2006-01-05 Nevidomski Alex Nevidomski Ale Method and system for approximate string matching
KR20080022481A (en) * 2006-09-06 2008-03-11 한미반도체 주식회사 Working table of pick and place system for semiconductor production
KR20080041471A (en) * 2006-11-07 2008-05-13 삼성전자주식회사 A die bonder
CN101290632A (en) * 2008-05-30 2008-10-22 北京搜狗科技发展有限公司 Input method for user words participating in intelligent word-making and input method system
CA2639919A1 (en) * 2007-10-01 2009-04-01 2012244 Ontario Inc. Handheld electronic device and associated method enabling prioritization of proposed spelling corrections
CN101727271A (en) * 2008-10-22 2010-06-09 北京搜狗科技发展有限公司 Method and device for providing error correcting prompt and input method system
JP2012049402A (en) * 2010-08-27 2012-03-08 Toshiba Corp Manufacturing device of semiconductor device, semiconductor device pickup method, and semiconductor device manufacturing method
WO2012167397A1 (en) * 2011-06-07 2012-12-13 Intel Corporation Dynamic soft keyboard for touch screen device
CN104412368A (en) * 2012-07-13 2015-03-11 应用材料公司 Method of diced wafer transportation
CN104600009A (en) * 2015-01-16 2015-05-06 圆融光电科技有限公司 Chip measuring sorting system and method
CN106469097A (en) * 2016-09-02 2017-03-01 北京百度网讯科技有限公司 A kind of method and apparatus recalling error correction candidate based on artificial intelligence
CN106708893A (en) * 2015-11-17 2017-05-24 华为技术有限公司 Error correction method and device for search query term
CN107305768A (en) * 2016-04-20 2017-10-31 上海交通大学 Easy wrongly written character calibration method in interactive voice
US20170364502A1 (en) * 2015-01-06 2017-12-21 What3Words Limited A Method For Suggesting Candidate Words As Replacements For An Input String Received At An Electronic Device
CN107704102A (en) * 2017-10-09 2018-02-16 北京新美互通科技有限公司 A kind of text entry method and device
CN111901665A (en) * 2020-08-28 2020-11-06 完美世界控股集团有限公司 Teaching resource playing method and device and storage medium
CN112560452A (en) * 2021-02-25 2021-03-26 智者四海（北京）技术有限公司 Method and system for automatically generating error correction corpus

Patent Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05175256A (en) * 1991-12-20 1993-07-13 Hitachi Ltd Pellet bonding device
US20060004744A1 (en) * 2004-06-19 2006-01-05 Nevidomski Alex Nevidomski Ale Method and system for approximate string matching
KR20080022481A (en) * 2006-09-06 2008-03-11 한미반도체 주식회사 Working table of pick and place system for semiconductor production
KR20080041471A (en) * 2006-11-07 2008-05-13 삼성전자주식회사 A die bonder
CA2639919A1 (en) * 2007-10-01 2009-04-01 2012244 Ontario Inc. Handheld electronic device and associated method enabling prioritization of proposed spelling corrections
CN101290632A (en) * 2008-05-30 2008-10-22 北京搜狗科技发展有限公司 Input method for user words participating in intelligent word-making and input method system
CN101727271A (en) * 2008-10-22 2010-06-09 北京搜狗科技发展有限公司 Method and device for providing error correcting prompt and input method system
JP2012049402A (en) * 2010-08-27 2012-03-08 Toshiba Corp Manufacturing device of semiconductor device, semiconductor device pickup method, and semiconductor device manufacturing method
WO2012167397A1 (en) * 2011-06-07 2012-12-13 Intel Corporation Dynamic soft keyboard for touch screen device
CN104412368A (en) * 2012-07-13 2015-03-11 应用材料公司 Method of diced wafer transportation
US20170364502A1 (en) * 2015-01-06 2017-12-21 What3Words Limited A Method For Suggesting Candidate Words As Replacements For An Input String Received At An Electronic Device
CN104600009A (en) * 2015-01-16 2015-05-06 圆融光电科技有限公司 Chip measuring sorting system and method
CN106708893A (en) * 2015-11-17 2017-05-24 华为技术有限公司 Error correction method and device for search query term
CN107305768A (en) * 2016-04-20 2017-10-31 上海交通大学 Easy wrongly written character calibration method in interactive voice
CN106469097A (en) * 2016-09-02 2017-03-01 北京百度网讯科技有限公司 A kind of method and apparatus recalling error correction candidate based on artificial intelligence
CN107704102A (en) * 2017-10-09 2018-02-16 北京新美互通科技有限公司 A kind of text entry method and device
CN111901665A (en) * 2020-08-28 2020-11-06 完美世界控股集团有限公司 Teaching resource playing method and device and storage medium
CN112560452A (en) * 2021-02-25 2021-03-26 智者四海（北京）技术有限公司 Method and system for automatically generating error correction corpus

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ANDUN6063: "基于海量词库的单词拼写检查、推荐到底 是咋做的？", pages 1 - 12, Retrieved from the Internet <URL:《https://blog.csdn.net/andun6063/article/details/101360211》> *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113743102A (en) * 2021-08-18 2021-12-03 百度在线网络技术（北京）有限公司 Method and device for recognizing characters and electronic equipment
CN113743102B (en) * 2021-08-18 2023-09-01 百度在线网络技术（北京）有限公司 Method and device for recognizing characters and electronic equipment

Also Published As

Publication number Publication date
CN112905026B (en) 2024-04-16

Similar Documents

Publication Publication Date Title
CN106534548B (en) Voice error correction method and device
CN109815314B (en) Intent recognition method, recognition device and computer readable storage medium
US7809664B2 (en) Automated learning from a question and answering network of humans
CN109299245B (en) Method and device for recalling knowledge points
US20050060301A1 (en) Question-answering method and question-answering apparatus
CN109033244B (en) Search result ordering method and device
CN106202153A (en) The spelling error correction method of a kind of ES search engine and system
US20100094855A1 (en) System for transforming queries using object identification
JPH0420220B2 (en)
CN104239565B (en) A kind of name automatic prompt method based on academics search
CN107807915B (en) Error correction model establishing method, device, equipment and medium based on error correction platform
US20220083577A1 (en) Information processing apparatus, method and non-transitory computer readable medium
CN109522397B (en) Information processing method and device
CN113722478B (en) Multi-dimensional feature fusion similar event calculation method and system and electronic equipment
CN112541095B (en) Video title generation method and device, electronic equipment and storage medium
CN107844531B (en) Answer output method and device and computer equipment
CN113239209A (en) Knowledge graph personalized learning path recommendation method based on RankNet-transformer
CN112667866A (en) Test paper generation method and device, electronic equipment and storage medium
WO2021239078A1 (en) Field recognition method, interaction method, electronic device, and storage medium
CN110262674A (en) Chinese character input method and device based on pinyin input and electronic equipment
CN112905026A (en) Method, device, storage medium and computer equipment for displaying word suggestions
JP5152918B2 (en) Named expression extraction apparatus, method and program thereof
CN111950267B (en) Text triplet extraction method and device, electronic equipment and storage medium
WO2024045926A1 (en) Multimedia recommendation method and recommendation apparatus, and head unit system and storage medium
CN112700203B (en) Intelligent marking method and device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant