WO2016045465A1 - 一种基于输入的信息展示方法和输入法*** - Google Patents

一种基于输入的信息展示方法和输入法*** Download PDF

Info

Publication number
WO2016045465A1
WO2016045465A1 PCT/CN2015/087058 CN2015087058W WO2016045465A1 WO 2016045465 A1 WO2016045465 A1 WO 2016045465A1 CN 2015087058 W CN2015087058 W CN 2015087058W WO 2016045465 A1 WO2016045465 A1 WO 2016045465A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
display information
display
input
semantic keyword
Prior art date
Application number
PCT/CN2015/087058
Other languages
English (en)
French (fr)
Inventor
李艳梅
李萍
Original Assignee
北京搜狗科技发展有限公司
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 北京搜狗科技发展有限公司 filed Critical 北京搜狗科技发展有限公司
Priority to US15/508,721 priority Critical patent/US10209782B2/en
Publication of WO2016045465A1 publication Critical patent/WO2016045465A1/zh

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
    • G06F3/0237—Character input methods using prediction or retrieval techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24575—Query processing with adaptation to user needs using context
    • 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
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis

Definitions

  • the present invention relates to the field of input method technologies, and in particular, to an input-based information display method and an input method system.
  • Input method refers to the encoding method used to input various symbols into a computer or other device, such as a mobile phone. If the user inputs by input method, the input method can convert the user's input content into corresponding text content, and perform on-screen according to user feedback.
  • the user also wants to insert the corresponding animation content such as an expression, a picture, etc.
  • the corresponding animation content such as an expression, a picture, etc.
  • the animation content that you want to input is on the screen, and the input process is cumbersome, which seriously affects the input efficiency.
  • the technical problem to be solved by the embodiments of the present invention is to provide an information display method based on input, which solves the problem that the input method text and the expression are separately input, and the efficiency is low.
  • an embodiment of the present invention further provides an input method device for ensuring implementation and application of the foregoing method.
  • the present invention discloses an input-based information display method, including: identifying and processing a received input character string by using context information, identifying a current semantic keyword; and determining a match with the semantic keyword.
  • the step of identifying the received input character string by using the context information, and identifying the current semantic keyword includes: converting the input character string, and identifying each of the input character strings a candidate; respectively combining the context information and each candidate into a candidate statement; performing semantic analysis on each candidate statement separately to identify the current semantic keyword.
  • the determining the display information that matches the semantic keyword comprises: matching locally stored display information with the semantic keyword, and determining display information that matches the semantic keyword; and/or And uploading the semantic keyword, and receiving the display information that is matched by the server and matched by the semantic keyword determined by the matching.
  • the method further includes: detecting whether an update data packet of the display information exists on the server side; and when there is an update data packet of the display information, Download the update package and store it locally.
  • the method further includes: identifying the current environment information, and filtering the display information that matches the semantic keyword according to the identified environment information.
  • the method further includes: sorting the display information according to a preset rule, and determining a sorting order of the display information, where the sorting factors in the preset rule include, but are not limited to, the number of times the display information is used. The time when the presentation information was last used, the recommendation index of the display information, and the degree of interest of the user.
  • the step of displaying the display information in the form of a bubble comprises: obtaining the first N display information according to the sorting order; and the front of the cloud in the form of a bubble in the cloud recommendation position of the current input method N display information for display.
  • the display information further includes: multimedia information and webpage link information.
  • the step of performing display processing on the display information according to the user instruction includes: when the display information is an expression image, selecting the expression image according to the user instruction, and selecting the expression The image is displayed on the screen; when the display information is multimedia information, the multimedia information is selected according to the user instruction, and the selected multimedia information is played; when the display information is webpage link information, The user instruction selects the webpage link information, and opens a webpage corresponding to the webpage link information.
  • the present invention also discloses an input method system, comprising: a keyword identification module, configured to identify and process a received input string by using context information, and identify a current semantic keyword; and display an information matching module. Determining the display information that matches the semantic keyword, wherein the display information includes but is not limited to an emoticon image; a bullet display module for displaying the display information in the form of a bullet; the display processing module, The display processing is performed according to a user instruction, where the display processing includes, but is not limited to, performing an upper screen operation on the expression image.
  • the keyword identification module is configured to convert the input character string, identify each candidate corresponding to the input character string, and combine the context information and each candidate into a candidate statement respectively Semantic analysis is performed separately for each candidate statement to identify the current semantic keyword.
  • the display information matching module is configured to match the locally stored display information with the semantic keyword to determine display information that matches the semantic keyword; and/or upload the semantic keyword And receiving display information that is matched by the server and matched by the semantic keyword determined by the matching.
  • the method further includes: an update module, configured to detect whether the update data packet of the display information exists on the server side; and when the update data packet of the display information exists, download the update data package and store the locality.
  • an update module configured to detect whether the update data packet of the display information exists on the server side; and when the update data packet of the display information exists, download the update data package and store the locality.
  • the method further includes: an environment identification module, configured to identify the current environment information, and filter the display information that matches the semantic keyword according to the identified environment information.
  • an environment identification module configured to identify the current environment information, and filter the display information that matches the semantic keyword according to the identified environment information.
  • the method further includes: a sorting module, configured to sort the display information according to a preset rule, and determine a sorting order of the display information, where the sorting factors in the preset rule include but are not limited to: The number of times the information is used, the time when the display information was last used, the recommendation index of the display information, and the degree of interest of the user.
  • a sorting module configured to sort the display information according to a preset rule, and determine a sorting order of the display information, where the sorting factors in the preset rule include but are not limited to: The number of times the information is used, the time when the display information was last used, the recommendation index of the display information, and the degree of interest of the user.
  • the bubble display module is configured to obtain the first N pieces of display information according to the sorting order; and display the first N pieces of display information in a form of a bullet in a cloud recommendation position of the current input method.
  • the display information further includes: multimedia information and webpage link information.
  • the display processing module is configured to: when the display information is an emoticon image, select the emoticon image according to the user instruction, and perform an upper screen on the selected emoticon image; When the multimedia information is used, the multimedia information is selected according to the user instruction, and the selected multimedia information is played; when the display information is the webpage link information, the webpage link information is performed according to the user instruction. Select and open the web page corresponding to the web page link information.
  • Embodiments of the present invention also disclose a program comprising readable code that, when executed on a computing device, causes the computing device to perform an input based operation according to any of the embodiments of the present invention Information display method.
  • the embodiment of the invention also discloses a readable medium, wherein the program as described in the embodiment of the invention is stored.
  • the embodiments of the invention include the following advantages:
  • the received input string is identified by the context information, so the identified current semantic keyword is not limited to the candidate corresponding to the input string, and the semantic recognition based on the context information is more in line with the user's input intention. And then determining the display information that matches the semantic keyword, so the display information is more in line with the input intention of the user, displaying the display information in the form of a bullet, and then displaying and displaying the display information according to the user instruction. If the display information includes an emoticon image, the emoticon image can be displayed on the screen according to the instruction information, so that the display information can be obtained and displayed before the text is not uploaded, and the input efficiency of the display information such as text and expression is improved.
  • FIG. 1 is a flow chart showing the steps of an embodiment of an input-based information display method according to the present invention
  • FIG. 2 is a flow chart showing the steps of an alternative embodiment of the input-based information display method of the present invention.
  • FIG. 3 is a structural block diagram of an embodiment of an input method system of the present invention.
  • FIG. 4 is a structural block diagram of an alternative embodiment of an input method system of the present invention.
  • Figure 5 shows a block diagram of a computing device for performing an input based information presentation method in accordance with the present invention
  • Figure 6 illustrates an input-based information display for implementing or carrying in accordance with the present invention.
  • the storage unit of the program code The storage unit of the program code.
  • One of the core concepts of the embodiments of the present invention is to provide an input-based information display method to solve the problem that the input method text and the expression are separately input, and the efficiency is low.
  • the received input string is identified by the context information, so the identified current semantic keyword is not limited to the candidate corresponding to the input string, and the semantic recognition based on the context information is more in line with the user's input intention.
  • determining the display information that matches the semantic keyword so the display information is more in line with the input intention of the user, displaying the display information in the form of a bullet, and then displaying and displaying the display information according to the user instruction.
  • the display information includes an emoticon image
  • the emoticon image can be displayed on the screen according to the instruction information, so that the display information can be obtained and displayed before the text is not uploaded, and the input efficiency of the display information such as text and expression is improved.
  • FIG. 1 a flow chart of steps of an embodiment of an input-based information display method according to the present invention is shown, which may specifically include the following steps:
  • Step 102 Identify the received input string by using context information, and identify the current semantic keyword.
  • the input method client can acquire the current input string, and transmit the input string to the kernel of the input method system, and the kernel analyzes and recognizes the received input string and a series of kernels.
  • Function such as inputting English characters to identify the corresponding English word candidates, and inputting Chinese characters to segment and convert the input characters to obtain corresponding Chinese candidates, and use context information for each function.
  • the candidate is further identified to identify the current semantic keyword.
  • the current input string is "gaoxing”
  • the candidate is identified as “happy”
  • the recorded context information is "I am not very today” so that the current semantic keyword is identified as "unhappy” based on the context information.
  • Another example is to get the current input string "bunanguode” and identify the candidate as “not sad. And the recorded context information is "I don't see today's situation”, so according to the context information, it is determined that the user wants to input is “not uncomfortable to see today's situation”, and the double negation is determined according to Chinese semantic analysis. Certainly, the semantic keyword is identified as "sad”.
  • the string “easy” is obtained, and the semantics of the problem is directly identified as “easy”.
  • the context information of the record is “take it”, and the context information determines that the user wants to input “take it easy”, and recognizes
  • the actual semantic keyword is "Easy.”
  • the input string is a user input content directly received by the input method system, and the user can input characters by using any input means, such as input through a keyboard, a touch screen, a voice, etc., such as through a tablet, a touch screen, a microphone, etc.
  • the input made.
  • the kernel refers to the core algorithm of the input method system, and the mechanism for outputting words according to the input content is reasonably displayed.
  • Step 104 Determine display information matching the semantic keyword, and display the display information in the form of a bullet.
  • the display information includes but is not limited to an expression image, that is, may be matched according to the semantic keyword.
  • Emoticon image The emoticon picture may be a static picture or a dynamic picture.
  • the input method may display the candidate corresponding to the input string, and display the display information corresponding to the semantic keyword in the form of a bullet, that is, previewing, recommending, and the like of the display information, so that the screen is not displayed before the screen is input.
  • the form of the bubble refers to a pop-up display box to display the form of the information, for example, the input method includes a button display area display button, a candidate display area display candidate, etc., in addition, a display box may be popped up to display based on Entering other information that matches, such as the display information matching the semantic keyword, that is, displaying the display information in the form of a bullet.
  • Step 106 Perform display processing on the display information according to a user instruction.
  • the display information When the display information is displayed in the form of the above-mentioned bullets, it can be regarded as a process of previewing and recommending the display information. After that, the user can select and process the information displayed by the bullet, so that after receiving the user instruction, the user can be based on the user.
  • the instruction selects the display information and performs display processing.
  • the display processing for the selected emoticon image may be packaged.
  • the method includes: performing, but not limited to, performing an upper screen on the emoticon image, so that in the character string input process, the corresponding display information is determined and displayed before the upper screen option is selected.
  • the candidate corresponding to the display information can be uploaded to the screen.
  • the corresponding keyword of the information displayed is “unhappy”, and the corresponding candidate is “happy”, so "Unhappy” corresponds to the emoticon image on the screen, the candidate is also "happy” on the screen; or only display the display information, and then select the candidate on the screen according to the user's instructions.
  • the above upper screen refers to the selected candidate of the input method output.
  • the candidate will be displayed in the chat after the screen is displayed.
  • the information input box of the terminal it will not be displayed in the chat window until it is confirmed to be sent.
  • the input method chat environment can be displayed on the screen to the information input box when the display information is on the screen, or can be uploaded to the chat window.
  • the semantic information that may affect the current display information may also be deleted from the context information that has been uploaded. Negative words and so on. For example, if the above input "gaoxing" identifies the display information determined by the semantic keyword as "unhappy", the negative context "No” may be deleted from the context information "I am not today", that is, the upper screen information is changed to "I am very good today” and then show the expression picture corresponding to "Unhappy".
  • the received input string is identified by the context information, so the identified current semantic keyword is not limited to the candidate corresponding to the input string, and the semantic recognition based on the context information is more in line with the user's input intention. . And then determining the display information that matches the semantic keyword, so the display information is more in line with the input intention of the user, displaying the display information in the form of a bullet, and then displaying and displaying the display information according to the user instruction. If the display information includes an emoticon image, the emoticon image can be displayed on the screen according to the instruction information, so that the display information can be obtained and displayed before the text is not uploaded, and the input efficiency of the display information such as text and expression is improved.
  • this embodiment further discusses a method of displaying the display information matching the input in the input process of the input method.
  • FIG. 2 a flow chart of an optional embodiment of an input-based information display method according to the present invention is shown. Specifically, the method may include the following steps:
  • Step 202 Convert the input character string to identify each candidate corresponding to the input character string.
  • Step 204 Combine the context information and each candidate into a candidate statement.
  • Step 206 Perform semantic analysis on each candidate statement separately to identify the current semantic keyword.
  • the user can obtain the current input string, and then transmit the current input string to the kernel of the input method system, and the kernel parses, segments, converts, recognizes, etc. the received input string. Processes the candidate corresponding to the input string.
  • the kernel records the option that the user just clicked on the screen, that is, the context information is recorded in advance. Therefore, when the kernel processes the input string, the kernel can also perform conversion, recognition, and the like according to the recorded context information, thereby obtaining more accurate. process result.
  • the candidate can be sorted according to the word frequency, user habits and the like. For example, when inputting English, the input string is recognized to obtain the corresponding English word candidate, and if the input string is input, the input string is divided and converted. Get the corresponding Chinese candidate.
  • the context information can be used to semantically analyze each candidate. Since the input string can match many candidates, it can be determined according to the requirements of which candidates are semantically analyzed, for example, only for the first screen candidate. Semantic analysis, such as semantic analysis of the top two candidates. Among them, the first screen candidate refers to the candidate displayed on the screen on the input method interface without page turning and sliding candidate operations.
  • the input method kernel recognizes the M candidates by input string conversion, and when the context information is used for analysis, the context information may be combined with each candidate corresponding to the input string to obtain corresponding M candidate statements. Therefore, each candidate statement is separately analyzed semantically, the meaning of the candidate is actually expressed in the sentence, and then the current semantic keyword is identified, so the semantic keyword may be the candidate or the slave statement. Information such as the words identified in the article.
  • the candidate sentences are segmented, and the nouns, verbs, adjectives, Keywords such as adverbs, prepositions, and auxiliary words can also be used to denoise the candidate sentences, such as deleting helpless words such as auxiliary words and prepositions, so as to determine the specific semantics of each keyword associated with the candidate according to the segmentation order, and obtain the semantic key. word.
  • the context information is "I don't see the situation today"
  • the input string identifies the candidate as “not sad” and “not sad” through the kernel, respectively forming the candidate statement A as "seeing today's situation” There is no unhappiness”; the candidate statement B is "not uncomfortable to see today's situation.”
  • the candidate statement A Taking the candidate statement A as an example, the words “see”, “today”, “of”, “situation”, “no”, “no”, “sad” and “” are deleted, and the preposition "of” is deleted.
  • the keywords related to the candidate “not sad” are “no”, “no” and “sad”. According to the order of the division, “nothing is not difficult”, according to the Chinese semantic double negation is affirmation, the semantic keyword is “ Sad.”
  • the context information includes the above information and the following information. If the current input string is located at the end of the document, that is, there is no document content thereafter, when the text input is simply performed, the above information of the current input string is obtained at this time. Corresponding if the current input string is at the forefront of the document, that is, there is no document content before it, then the following information of the current input string is obtained; if the current input string is located in the middle of the document, such as editing and modifying the document , the document content exists before and after, and the above information and the following information are obtained at the same time.
  • the candidate ranked in the first two is combined with the context information.
  • the input string is “kaixin”, and the first two candidates are “happy” and “open new”.
  • the above information is “I”. Now it is very non-”, then the candidate statement is "I am very unhappy now”, and "I am very open now”, each of the candidate statements are semantically analyzed, such as through the semantics of segmentation, filtering, denoising, etc.
  • the semantic keyword determined by "I am very unhappy now” is "unhappy”, so that based on the context information, the keyword "unhappy” which is more in line with the user's input intention can be matched, if only the input string is followed. If the candidate is determined to be "happy”, it is likely to deviate from the user's input intention and get a wrong recognition result, such as recommending a happy expression when the user is unhappy, affecting the user's experience.
  • Step 208 Detect whether an update data packet of the display information exists on the server side.
  • the terminal can detect whether the terminal is connected to the network. If the terminal is connected to the network, it can detect whether there is an update data packet showing the information on the server side.
  • the update data packet refers to the display letter generated within a certain time.
  • the data packet is generated after the data packet of the input method last synchronized display information, that is, the data packet detecting whether the presence information is present on the server side is not stored locally.
  • the recommendation index may be set according to actual needs, for example, according to the heat calculation of the overall user's use of the new expression, Such as the time generated by the expression and other information.
  • step 210 is performed; if no, that is, the update data packet of the display information does not exist on the server side, step 212 is performed.
  • Step 210 Download the update data packet and store it locally.
  • the update data packet of the display information exists, the update data packet is downloaded and stored locally, that is, the data packet of the locally stored display information is synchronized, so that the display information can be directly matched locally to improve matching. Efficiency, and can ensure that even if the input method client is not connected to the network, it can obtain newer display information and provide recommendation information that meets the user's intention.
  • Step 212 Determine display information that matches the semantic keyword.
  • the determining the display information that matches the semantic keyword comprises: matching the locally stored display information with the semantic keyword, and determining the display that matches the semantic keyword. Information; and/or, uploading the semantic keyword, and receiving display information that is matched by the server and matched by the semantic keyword determined by the matching.
  • the display information corresponding to “happy” may be a smiley face, or may be a picture of a cartoon character jumping, so that the display information set may be used to store the same type of display information.
  • the semantic keyword and the display information are not strictly one-to-one correspondence, different semantic keywords may correspond to the same display information set, and there may be multiple display information in one display information set, such as “unhappy” and “unhappy” "may correspond to the same set of presentation information.
  • the display information includes but is not limited to: an emoticon image, a multimedia information, and a webpage link information
  • the multimedia information includes audio, video, flash animation, and the like
  • the webpage link information may be various information such as news, encyclopedia, and translation.
  • the semantic keyword can be matched with the locally stored display information, or the semantic keyword can be uploaded, thereby adopting the storage exhibition on the server side.
  • the information is matched to the semantic keyword, or both are matched, so that the display information matching the semantic keyword is obtained, wherein the display information may be all display information in one or more collections.
  • the emoticon image can be matched with the locally stored display information, and the matching display information can be obtained from the server side for the webpage link information, and the multimedia information can be matched both locally and on the server side.
  • the semantic keyword "unhappy” can match 5 emoticons, and the corresponding images of the emoticons are A1 to A5.
  • Step 214 Identify the current environment information, and filter the display information matching the semantic keyword according to the identified environment information.
  • the emoticon picture may be a static emoticon picture of .jpg or a dynamic emoticon picture of the GIF; if the current environment information is a short message editing Environment, emoticons can support emoji expressions and character-like expressions. For multimedia information, it may not be displayed in the SMS editing environment.
  • Step 216 Sort the display information according to a preset rule, and determine a sort order of the display information.
  • the ordering rules of the display information are preset, and the display information filtered by the environment information may be sorted according to the preset rules, and the sorting order of each display information is determined.
  • the preset rule may be configured with multiple sorting factors, including but not limited to: the number of times the user uses the display information, the time when the display information was last used, the recommendation index of the display information, and the user's interest level.
  • sorting may be based on one or more sorting factors, for example, when sorting based on a plurality of sorting factors, different weighting information may be configured without each sorting factor, thereby performing weighted summation based on the respective weight information, and then Sort according to the calculation results, determine the display letter The sort order of the information.
  • the number of times the display information is used refers to the number of times the current user uses the display information.
  • the time when the display information was last used refers to the time interval when the current user last used the current time, and the time interval can be determined in units of days and minutes.
  • the recommendation index of the display information refers to the recommendation degree that the server recommends the display information to the user, and the server can comprehensively determine the recommendation index according to the generation time of the display information, the download amount of the user, and the actual content of the display information, for example, an expression.
  • the recommendation index of the expression image related to the holiday during the holiday is higher, and, for example, when the web news is used as the display information, the news recommendation index with a relatively late time or a significant event is higher.
  • the degree of interest of the user refers to the degree of preference of the user for the displayed information, for example, the number of users who can use, view, and evaluate by statistics.
  • the user's usage habits and recommendation indexes may be comprehensively calculated.
  • the calculation principle is as follows: coefficient 1 (for example, 0.6) * frequency used by the current user + coefficient 2 ( For example, 0.2)* (the highest emoticon frequency used * the current emoticon function used by the current emoticon to the present time - the frequency used by the current emoticon user) + coefficient 3 (for example, 0.2) * recommended index. Therefore, the display information can be sorted according to the calculation result.
  • the time interval probability function is a probability function obtained by analyzing the screen condition and time relationship data of the expression of the overall user.
  • the semantic keyword "unhappy” can match 5 emoticons
  • the corresponding embedding codes are A1 to A5
  • A5 is the recommended emoticon, which can calculate A1-A5 according to user habits and recommendation index.
  • Determine the corresponding sort value to sort, the sort order is (A2, A3, A5, A4, A1).
  • step 218 the display information is displayed in the form of a bullet.
  • the displaying the display information in the form of a bullet comprising: obtaining the first N display information according to the sorting order; and the cloud recommendation position of the current input method is in the form of a bullet
  • the first N display information is displayed.
  • the first N pieces of display information are displayed in the form of a bubble in the cloud recommendation position of the input method keyboard, that is, the display frame of the bullet is displayed at the cloud recommendation position of the input method keyboard, and then the front N is displayed in the display frame. Display information.
  • the sort order is (A2, A3, A5, A4, A1), and the first two emoticons are selected as A2 and A3.
  • Step 220 Perform display processing on the display information according to a user instruction.
  • the user can select to display the candidate, the display information, etc., and thus can receive the user instruction, and then display and display the display information according to the user instruction.
  • the display information includes a variety of different information, and the display manner of different information also has certain differences, which are discussed in detail below.
  • the emoticon image is selected according to the user instruction, and the selected emoticon image is displayed on the screen. That is, the expression image displayed in the form of the bubble can be selected according to the user instruction, and then the selected expression image is subjected to the upper screen processing, that is, the expression image is displayed in the information input area of the current input environment, wherein the expression image can be corresponding to the candidate Items are displayed together on the screen.
  • the screen is displayed, it is displayed in the same information box or in different display boxes, or only the image is displayed on the screen.
  • the candidates can be processed according to the subsequent instructions.
  • the display information is multimedia information
  • selecting the multimedia information according to the user instruction, and playing the selected multimedia information the corresponding player may be started to play the multimedia information, or may be based on the current
  • the environment information of the application determines how to display the multimedia information, such as playing directly in the application, and for the instant messaging application, the multimedia information can also be sent to the friend (information receiver) for recommendation.
  • the webpage link information is selected according to the user instruction, and the webpage corresponding to the webpage link information is opened, and an application program such as a browser may be invoked to open the webpage corresponding to the webpage link information. It is also possible to determine how to display and process the webpage link information according to the current environment information of the application. For example, for the instant messaging application, the webpage link information may be sent to the friend and recommended to the friend for viewing.
  • the candidate corresponding to the input string is combined with the context information to identify the display information that meets the user's input intention. And based on multiple sorting factors in the preset rules Sort the display information to get the impression information that the user is most likely to choose.
  • the embodiment displays the display information on the cloud recommendation position, which can avoid occupying the position of the first screen candidate, reduce the operation steps of the user to find the recommended information and the candidate, and improve the input efficiency.
  • the embodiment further provides an input method system.
  • FIG. 3 a structural block diagram of an embodiment of an input method system of the present invention is shown, which may specifically include the following modules:
  • the keyword identification module 302 is configured to identify and process the received input character string by using context information, and identify the current semantic keyword.
  • the display information matching module 304 is configured to determine display information that matches the semantic keyword, wherein the display information includes but is not limited to an expression image.
  • the bubble display module 306 is configured to display the display information in the form of a bullet.
  • the display processing module 308 is configured to perform display processing on the display information according to a user instruction, where the display processing includes, but is not limited to, performing an upper screen operation on the expression image.
  • the received input string is identified by the context information, so the identified current semantic keyword is not limited to the candidate corresponding to the input string, and the semantic recognition based on the context information is more in line with the user's input intention. . And then determining the display information that matches the semantic keyword, so the display information is more in line with the input intention of the user, displaying the display information in the form of a bullet, and then displaying and displaying the display information according to the user instruction. If the display information includes an emoticon image, the emoticon image can be displayed on the screen according to the instruction information, so that the display information can be obtained and displayed before the text is not uploaded, and the input efficiency of the display information such as text and expression is improved.
  • FIG. 4 a structural block diagram of an optional embodiment of an input method system according to the present invention is shown, which may specifically include the following modules:
  • the keyword identification module 302 is configured to convert the input character string, identify each candidate corresponding to the input character string, and combine the context information and each candidate into a candidate statement respectively; The candidate statements are separately analyzed for semantics to identify the current semantic keywords.
  • the display information matching module 304 is configured to match the locally stored display information with the semantic keyword to determine display information that matches the semantic keyword; and/or upload the semantic keyword and receive The display information that is matched by the server and matched by the semantic keyword determined by the matching.
  • the input method system also includes:
  • the update module 310 is configured to detect whether the update data packet of the display information exists on the server side, and download the update data packet and store it locally when the update data packet of the display information exists.
  • the environment identification module 312 is configured to identify the current environment information, and filter the display information that matches the semantic keyword according to the identified environment information.
  • the sorting module 314 is configured to sort the display information according to a preset rule, and determine a sorting order of the display information, where the sorting factors in the preset rule include, but are not limited to, the number of times the display information is used, The time when the presentation information was last used, the recommendation index of the presentation information, and the degree of interest of the user.
  • the bullet display module 306 is configured to obtain the first N pieces of display information according to the sorting order; and display the top N pieces of display information in a form of a bullet in the cloud recommendation position of the current input method.
  • the display information further includes: multimedia information and webpage link information.
  • the display processing module 308 is configured to: when the display information is an emoticon image, select the emoticon image according to the user instruction, and perform an upper screen on the selected emoticon image; and when the display information is multimedia information And selecting, according to the user instruction, the multimedia information, and playing the selected multimedia information; when the display information is webpage link information, selecting the webpage link information according to the user instruction, and Open the web page corresponding to the web page link information.
  • the candidate corresponding to the input string is combined with the context information to identify the display information that meets the user's input intention.
  • the display information is sorted according to multiple sorting factors in the preset rule, and the display information that the user is most likely to select is obtained.
  • the embodiment displays the display information on the cloud recommendation position, which can avoid occupying the position of the first screen candidate, reduce the operation steps of the user to find the recommended information and the candidate, and improve the input efficiency.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
  • embodiments of the embodiments of the invention may be provided as a method, apparatus, or computer program product.
  • embodiments of the invention may be in the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware.
  • embodiments of the invention may take the form of a computer program product embodied on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • Embodiments of the invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal device to produce a machine such that instructions are executed by a processor of a computer or other programmable data processing terminal device
  • Means are provided for implementing the functions specified in one or more of the flow or in one or more blocks of the flow chart.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • Instruction device implemented in a flowchart A function specified in a block or blocks of a process or multiple processes and/or block diagrams.
  • Figure 5 illustrates a computing device that can implement the adjustment of the input method virtual keyboard in accordance with the present invention.
  • the computing device conventionally includes a processor 510 and a program product or readable medium in the form of a memory 520.
  • Memory 520 can be an electronic memory such as a flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), EPROM, or ROM.
  • Memory 520 has a memory space 530 for program code 531 for performing any of the method steps described above.
  • storage space 530 for program code may include various program code 531 for implementing various steps in the above methods, respectively.
  • These program codes can be read from or written to one or more program products.
  • These program products include program code carriers such as memory cards.
  • Such a program product is typically a portable or fixed storage unit as described with reference to FIG.
  • the storage unit may have storage segments, storage spaces, and the like that are similarly arranged to memory 520 in the computing device of FIG.
  • the program code can be compressed, for example, in an appropriate form.
  • the storage unit includes readable code 531', ie, code that can be read by a processor, such as 510, that when executed by a computing device causes the computing device to perform various steps in the methods described above .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种基于输入的信息展示方法和输入法装置，以解决输入法文字和表情分开输入，效率较低的问题。所述的方法包括：采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词（102）；确定与所述语义关键词匹配的展示信息，并以弹泡形式对所述展示信息进行展示（104），其中，所述展示信息包括但不限于表情图片；依据用户指令对所述展示信息进行显示处理（106），其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。该方法和装置识别出的当前的语义关键词不限于输入字符串对应的候选项，且基于上下文信息的语义识别更加符合用户的输入意图，展示信息也更加符合用户的输入意图，提高文字和表情等展示信息的输入效率。

Description

一种基于输入的信息展示方法和输入法***
本申请要求申请号为201410499401.6、申请日为2014年9月25日、发明名称为“一种基于输入的信息展示方法和输入法***”的发明专利申请的全部优先权。
技术领域
本发明涉及输入法技术领域，特别是涉及一种基于输入的信息展示方法和一种输入法***。
背景技术
输入法是指为将各种符号输入计算机或其他设备(如手机)而采用的编码方法。如用户采用输入法进行输入时，输入法可以将用户的输入内容转换成相应的文字内容，并依据用户反馈进行上屏。
但有时用户输入文字的过程中，还想要***相应的表情、图片等动画内容，此时就要反馈要上屏的文字内容并在该文字内容上屏后，再进入动画的相关插件中选择想要输入的动画内容进行上屏，输入过程比较繁琐，严重影响输入效率。
发明内容
本发明实施例所要解决的技术问题是提供一种基于输入的信息展示方法，以解决输入法文字和表情分开输入，效率较低的问题。
相应的，本发明实施例还提供了一种输入法装置，用以保证上述方法的实现及应用。
为了解决上述问题，本发明公开了一种基于输入的信息展示方法，包括：采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词；确定与所述语义关键词匹配的展示信息，并以弹泡形式对所述展示信息进行展示，其中，所述展示信息包括但不限于表情图片；依据用户指令对所述展示信息进行显示处理，其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。
可选的，所述采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词的步骤，包括：对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项；分别将所述上下文信息和每个候选项组合成候选语句；对每个候选语句分别进行语义分析，识别出当前的语义关键词。
可选的，所述确定与所述语义关键词匹配的展示信息，包括：将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或，上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
可选的，所述确定与所述语义关键词匹配的展示信息的步骤之前，还包括：检测服务器侧是否存在所述展示信息的更新数据包；当存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地。
可选的，确定与所述语义关键词匹配的展示信息的步骤之后，还包括：对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
可选的，还包括：按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序，其中所述预置规则中的排序因素包括但不限于：所述展示信息的使用次数、上次使用所述展示信息的时间、所述展示信息的推荐指数和用户的兴趣度。
可选的，所述以弹泡形式对所述展示信息进行展示的步骤，包括：按照所述排序顺序获取前N个展示信息；在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
可选的，所述展示信息还包括：多媒体信息和网页链接信息。
可选的，所述依据用户指令对所述展示信息进行显示处理的步骤，包括：当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏；当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对选择的多媒体信息进行播放；当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页。
相应的，本发明还公开了一种输入法***，包括：关键词识别模块，用于采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词；展示信息匹配模块，用于确定与所述语义关键词匹配的展示信息，其中，所述展示信息包括但不限于表情图片；弹泡展示模块，用于以弹泡形式对所述展示信息进行展示；显示处理模块，用于依据用户指令对所述展示信息进行显示处理，其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。
可选的，所述关键词识别模块，用于对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项；分别将所述上下文信息和每个候选项组合成候选语句；对每个候选语句分别进行语义分析，识别出当前的语义关键词。
可选的，所述展示信息匹配模块，用于将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或，上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
可选的，还包括：更新模块，用于检测服务器侧是否存在所述展示信息的更新数据包；并在存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地。
可选的，还包括：环境识别模块，用于对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
可选的，还包括：排序模块，用于按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序，其中所述预置规则中的排序因素包括但不限于：所述展示信息的使用次数、上次使用所述展示信息的时间、所述展示信息的推荐指数和用户的兴趣度。
可选的，所述弹泡展示模块，用于按照所述排序顺序获取前N个展示信息；在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
可选的，所述展示信息还包括：多媒体信息和网页链接信息。
可选的，所述显示处理模块，用于当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏；当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对选择的多媒体信息进行播放；当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页。
本发明实施例还公开了一种程序，包括可读代码，当所述可读代码在计算设备上运行时，导致所述计算设备执行根据本发明实施例中的任一个所述的基于输入的信息展示方法。
本发明实施例还公开了一种可读介质，其中存储了如本发明实施例中所述的程序。
与现有技术相比，本发明实施例包括以下优点：
采用上下文信息对接收的输入字符串进行识别处理，因此识别出的当前的语义关键词不限于输入字符串对应的候选项，而基于上下文信息的语义识别更加符合用户的输入意图。然后确定与所述语义关键词匹配的展示信息，因此该展示信息也更加符合用户的输入意图，以弹泡形式对所述展示信息进行展示，再依据用户指令对所述展示信息进行显示处理，如展示信息包括表情图片，则可以依据指令信息对表情图片进行上屏，从而可以在文字未上屏前就获取展示信息并展示，提高文字和表情等展示信息的输入效率。
附图说明
图1是本发明的一种基于输入的信息展示方法实施例的步骤流程图；
图2是本发明的一种基于输入的信息展示方法可选实施例的步骤流程图；
图3是本发明一种输入法***实施例的结构框图；
图4是本发明一种输入法***可选实施例的结构框图；
图5示出了用于执行根据本发明的基于输入的信息展示方法的计算设备的框图；
图6示出了用于保持或者携带实现根据本发明的基于输入的信息展示方 法的程序代码的存储单元。
具体实施方式
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合附图和具体实施方式对本发明作进一步详细的说明。
本发明实施例的核心构思之一在于，提供一种基于输入的信息展示方法，以解决输入法文字和表情分开输入，效率较低的问题。采用上下文信息对接收的输入字符串进行识别处理，因此识别出的当前的语义关键词不限于输入字符串对应的候选项，而基于上下文信息的语义识别更加符合用户的输入意图。然后确定与所述语义关键词匹配的展示信息，因此该展示信息也更加符合用户的输入意图，以弹泡形式对所述展示信息进行展示，再依据用户指令对所述展示信息进行显示处理，如展示信息包括表情图片，则可以依据指令信息对表情图片进行上屏，从而可以在文字未上屏前就获取展示信息并展示，提高文字和表情等展示信息的输入效率。
实施例一
参照图1，示出了本发明的一种基于输入的信息展示方法实施例的步骤流程图，具体可以包括如下步骤：
步骤102，采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词。
在用户采用输入法进行输入的过程中，输入法客户端可以获取当前输入字符串，并将输入字符串传输给输入法***的内核，内核对接收的输入字符串进行解析、识别等一系列内核功能，如输入英文时对输入字符串进行识别获取其对应的英文单词候选项，又如输入中文时对输入字符串进行切分、转化等操作获取对应的中文候选项，并采用上下文信息对各候选项进行进一步识别，从而识别出当前的语义关键词。
例如，获取当前输入字符串为“gaoxing”，识别出候选项为“高兴”，并且记录的上下文信息为“我今天很不”，从而依据上下文信息识别出当前的语义关键词为“不高兴”。
又如，获取当前输入字符串“bunanguode”，识别出候选项为“不难过 的”，并且记录的上下文信息为“看到今天的情况没有”，从而依据上下文信息确定用户想要输入的是“看到今天的情况没有不难过的”，根据中文语义分析确定双重否定即为肯定，则识别出语义关键词为“难过”。
又如，获取字符串“easy”，直接识别问题的语义为“容易”，但是，记录的上下文信息为“take it”，依据上下文信息确定用户想要输入的是“take it easy”，识别出实际的语义关键词为“放轻松”。
其中，输入字符串是输入法***直接接收的用户输入内容，用户可以采用任意的输入手段进行字符的输入，如通过键盘、触摸屏、语音等进行的输入，又如通过手写板、触摸屏、麦克风等进行的输入。内核指的是输入法***的核心算法，根据输入内容合理展现输出词语的机制。
步骤104，确定与所述语义关键词匹配的展示信息，并以弹泡形式对所述展示信息进行展示。
按照识别出的语义关键词从本地或网络侧匹配展示信息，确定所述语义关键词对应的展示信息，本实施例中，所述展示信息包括但不限于表情图片，即可以依据语义关键词匹配表情图片。其中，表情图片可以是静态图片也可以是动态图片。
输入法可以对上述输入字符串对应的候选项进行展示，并且以弹泡的形式展示语义关键词对应的展示信息，即对展示信息进行预览、推荐等，从而在文字输入过程中未上屏前就匹配出展示信息。其中，弹泡的形式指的是弹出展示框来显示信息的形式，例如输入法中包括按键显示区域显示按键，候选展示区域显示候选项等，在此之外，还可以弹出展示框来显示基于输入匹配的其他信息，如所述与所述语义关键词匹配的展示信息，即以弹泡形式对所述展示信息进行展示。
步骤106，依据用户指令对所述展示信息进行显示处理。
上述弹泡形式展示所述展示信息时，可以看作是对展示信息的预览、推荐等处理，此后用户可以对弹泡展示的信息进行选择等处理，从而在接收到用户指令后，可以依据用户指令对展示信息进行选择并执行显示处理。
例如，展示信息为表情图片时，对于选择的表情图片的显示处理可以包 括但不限于：对所述表情图片进行上屏，从而在字符串输入过程中，未选择上屏选项前就确定对应的展示信息并显示。
并且在显示处理时，可以将与该展示信息对应的候选项进行上屏，如上例中展示信息对应义关键词为“不高兴”，其对应的候选项为“高兴”，因此可以在将“不高兴”对应表情图片上屏时，也将候选项为“高兴”上屏；或者仅显示该展示信息，后续再依据用户的指示选择候选项上屏。
上述上屏指的是输入法输出选定的候选项，其中，依据当前应用环境不同上屏也存在多种不同，例如在即时通讯应用的聊天环境中，候选项上屏后会展示在聊天本端的信息输入框中，确认发送后才会显示到聊天窗口中。而输入法聊天环境对展示信息进行上屏时可以上屏到信息输入框，也可以上屏到聊天窗口中。
并且，由于依据语义关键词确定的展示信息可能与当前输入候选项的语义不同，因此在选择展示信息进行上屏后，还可以从已上屏的上下文信息中删除可能影响当前展示信息的语义的否定词等内容。例如对上述输入“gaoxing”识别语义关键词为“不高兴”所确定的展示信息，可以将已上屏的上下文信息“我今天很不”中删除否定词“不”，即上屏信息更改为“我今天很”然后展示“不高兴”对应的表情图片。
综上所述，采用上下文信息对接收的输入字符串进行识别处理，因此识别出的当前的语义关键词不限于输入字符串对应的候选项，而基于上下文信息的语义识别更加符合用户的输入意图。然后确定与所述语义关键词匹配的展示信息，因此该展示信息也更加符合用户的输入意图，以弹泡形式对所述展示信息进行展示，再依据用户指令对所述展示信息进行显示处理，如展示信息包括表情图片，则可以依据指令信息对表情图片进行上屏，从而可以在文字未上屏前就获取展示信息并展示，提高文字和表情等展示信息的输入效率。
实施例二
在上述实施例的基础上，本实施例进一步论述在输入法的输入过程中展示与输入匹配的展示信息的方法。
参照图2，示出了本发明的一种基于输入的信息展示方法可选实施例的步骤流程图，具体可以包括如下步骤：
步骤202，对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项。
步骤204，分别将所述上下文信息和每个候选项组合成候选语句。
步骤206，对每个候选语句分别进行语义分析，识别出当前的语义关键词。
用户在采用输入法进行输入的过程中，可以获取当前的输入字符串，然后对当前输入字符串传输给输入法***的内核，内核对接收的输入字符串进行解析、切分、转换、识别等处理获取输入字符串对应的候选项。其中，内核会对用户刚上屏的选项进行记录，即预先记录上下文信息，因此内核在对该输入字符串处理时，还可以依据记录的上下文信息进行转换、识别等处理，从而得到更准确的处理结果。并可以按照词频、用户习惯等信息对候选项进行排序，如输入英文时对输入字符串进行识别获取其对应的英文单词候选项，又如输入中文时对输入字符串进行切分、转化等操作获取对应的中文候选项。
在获取到各候选项之后，可以采用上下文信息对各候选项进行语义分析，由于输入字符串可以匹配出很多候选项，可以按照需求确定对哪些候选项进行语义分析，例如仅对首屏候选项进行语义分析，又如对排在前两位的候选项进行语义分析等。其中，首屏候选项指的是在输入法界面上，不进行翻页及滑动候选操作即可展示在屏幕上的候选项。
例如输入法内核将输入字符串转换识别出M个候选项，在采用上下文信息进行分析时，可以将上下文信息分别与所述输入字符串对应每个候选项进行组合，获取相应的M个候选语句，从而对每个候选语句分别进行语义分析，确定该候选项在语句中实际表述的含义，进而确定识别出当前的语义关键词，因此该语义关键词可能是该候选项，也可能是从语句中识别出的词条等信息。
例如对候选语句进行切分，切分得到候选语句中的名词、动词、形容词、 副词、介词、助词等关键词，还可以对候选语句进行去噪处理，如删除助词、介词等无用词，从而按照切分顺序针对与候选项关联的各关键词确定其具体语义，得到语义关键词。
如上例中，上下文信息为“看到今天的情况没有”，输入字符串通过内核识别出候选项为“不难过的”和“不难过地”，分别构成候选语句A为“看到今天的情况没有不难过的”；候选语句B为“看到今天的情况没有不难过地”。以候选语句A为例，切分的到“看到”、“今天”、“的”、“情况”、“没有”、“不”、“难过”和“的”，删除介词“的”，与候选项“不难过的”相关的关键词是“没有”、“不”和“难过”，按照切分顺序为“没有不难过”，依据中文语义双重否定即为肯定，语义关键词为“难过”。
本实施例中，上下文信息包括上文信息和下文信息，若当前输入字符串位于文档的最后，即其后无文档内容，单纯进行文字输入时，则此时获取当前输入字符串的上文信息；对应的若当前输入字符串位于文档的最前端，即其前无文档内容，则此时获取当前输入字符串的下文信息；若当前输入字符串位于文档的中间，如对文档进行编辑修改时，其前后均存在文档内容，则同时获取其上文信息和下文信息。
例如将排在前两位的候选项与上下文信息进行结合，例如，输入字符串为“kaixin”，排在前两位的候选项分别为“开心”“开新”，上文信息为“我现在非常不”，则结合后得到候选语句为“我现在非常不开心”，以及“我现在非常不开新”，分别每个候选语句进行语义分析，如通过切分、筛选、去噪等语义分析处理后候选语句为“我现在非常不开心”确定的语义关键词为“不开心”，从而基于上下文信息可以匹配到更符合用户输入意图的关键词“不开心”，若仅按照输入字符串的候选项确定“开心”，则很有可能和用户的输入意图背离，得到错误的识别结果，如在用户不开心的情况下推荐了开心的表情，影响用户的体验效果。
步骤208，检测服务器侧是否存在所述展示信息的更新数据包。
可以检测终端是否连接网络，若终端连接网络则可以检测服务器侧是否存在展示信息的更新数据包，更新数据包指的是在一定时间内生成的展示信 息的数据包，如在输入法上次同步展示信息的数据包之后生成的，即检测服务器侧是否存在展示信息的数据包在本地未存储。
其中，由于服务器侧可能存在很多推荐的更新数据包，此时可以按照推荐指数确定下载哪些数据包，其中，推荐指数可以根据实际需求设定，例如根据整体用户对新表情使用的热度计算，又如根据表情生成的时间等信息。
若是，即服务器侧存在所述展示信息的更新数据包，执行步骤210；若否，即服务器侧不存在所述展示信息的更新数据包，执行步骤212。
步骤210，下载所述更新数据包并存储到本地。
当存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地，即对本地存储的展示信息的数据包进行同步，从而在本地即可直接对展示信息进行匹配，提高匹配的效率，并且可以确保即使输入法客户端未联网也能获取比较新的展示信息，提供符合用户意图的推荐信息。
步骤212，确定与所述语义关键词匹配的展示信息。
本发明一个可选实施例中，所述确定与所述语义关键词匹配的展示信息，包括：将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或，上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
本实施例中，由于同一种类的展示信息可能存在多个，如“高兴”对应的展示信息可能是笑脸，也可能是卡通人物蹦跳的图片等，因此可以采用展示信息集合存储同一种类的展示信息，并且语义关键词和展示信息并不是严格的一一对应关系，不同语义关键词可能对应相同的展示信息集合，一个展示信息集合中可能存在多个展示信息，例如“不高兴”和“不开心”可能对应相同的展示信息集合。
并且，展示信息包括但不限于：表情图片、多媒体信息和网页链接信息等，多媒体信息包括音、视频，Flash动画等信息，网页链接信息可以是新闻、百科、翻译等各类信息。
在采用语义关键词匹配展示信息时，可以将语义关键词与本地存储的展示信息进行匹配，也可以上传该语义关键词，从而在服务器侧采用存储的展 示信息对该语义关键词进行匹配，或者两者均进行匹配，从而获取到与所述语义关键词匹配的展示信息，其中该展示信息可能是一个或多个集合内的所有展示信息。
例如对于表情图片可以与本地存储的展示信息进行匹配，对于网页链接信息可以从服务器侧获取匹配的展示信息，对于多媒体信息可以在本地和服务器侧均进行匹配等。
例如上例中语义关键词“不开心”可以匹配出5个表情图片，该表情图片对应的编码分别为A1至A5。
步骤214，对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
由于不同的环境信息对展示信息具有不同的要求，因此还要识别出当前的环境信息，然后依据环境信息对匹配出的展示信息进行筛选，其中环境信息可以用于对具体的展示信息进行筛选，也可以对展示信息的类型进行筛选。通过环境信息可以筛选出符合当前输入环境的展示信息，从而避免出现展示信息无法显示等问题。
例如，对于表情图片而言，若当前的环境信息为即时通信客户端等在线交互环境，则表情图片可以是.jpg的静态表情图片，或者是GIF的动态表情图片；若当前环境信息为短信编辑环境，则表情图片可以使支持emoji表情和字符类表情等。而对于多媒体信息可能在短信编辑环境中不进行显示。
步骤216，按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序。
本实施例中预置了展示信息的排序规则，可以按照预置规则对依据环境信息筛选出的展示信息进行排序，确定各展示信息的排序顺序。其中，预置规则中可以配置多种排序因素，该排序因素包括但不限于：用户对展示信息的使用次数、上次使用该展示信息的时间、所述展示信息的推荐指数以及用户的兴趣度等，可以基于其中的一种或多种排序因素进行排序，例如基于多种排序因素进行排序时，可以未每种排序因素配置不同的权重信息，从而基于各自的权重信息进行加权求和，然后按照计算结果进行排序，确定展示信 息的排序顺序。
其中，展示信息的使用次数指的是当前用户对该展示信息的使用次数。
上次使用该展示信息的时间指的是当前用户上一次使用到当前时间的时间间隔，可以按照天、分钟为单位确定时间间隔。
展示信息的推荐指数指的是服务器将该展示信息推荐给用户使用的推荐程度，服务器可以依据该展示信息的生成时间、用户的下载量，展示信息的实际内容等综合确定推荐指数，例如，表情图片作为展示信息时，在节假日期间与该节假日相关表情图片的推荐指数较高，又如，网页新闻作为展示信息时，发生时间较近或事件重大的新闻推荐指数较高。
用户的兴趣度指的是用户对该展示信息的喜好程度，例如可以通过统计使用、查看、评价的用户数量确定。
例如按照预置规则中的排序因素对展示信息进行排序时，可以对用户使用习惯和推荐指数等进行综合计算，计算原理大致如下：系数1(例如0.6)*当前用户使用的频次+系数2(例如0.2)*(被使用最高的表情频次*当前表情上次使用到现在的时间间隔概率函数-当前表情用户使用的频次)+系数3(例如0.2)*推荐的指数。从而可以依据该计算结果对该展示信息进行排序。
其中，时间间隔概率函数是通过分析整体用户的表情上屏情况和时间关系数据得到的概率函数。
如上例中语义关键词“不开心”可以匹配出5个表情图片，该表情图片对应的编码分别为A1至A5，A5为推荐的表情，可以根据用户使用习惯和推荐指数对A1-A5进行计算确定相应的排序值进行排序，排序顺序为(A2，A3、A5、A4、A1)。
步骤218，以弹泡形式对所述展示信息进行展示。
本发明一个可选实施例中，所述以弹泡形式对所述展示信息进行展示，包括：按照所述排序顺序获取前N个展示信息；在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
上述对展示信息进行排序后，可以按照该排序顺序确定排在前N位展示信息，其中，N为正整数，如取N＝2，则获取前两个展示信息，然后可以对 前N个展示信息进行展示。例如在输入法键盘的云推荐位置以弹泡形式对所述前N个展示信息进行展示，即在输入法键盘的云推荐位置处显示弹泡的展示框，然后在该展示框中展示前N个展示信息。
如上例中排序顺序为(A2，A3、A5、A4、A1)，从中选取前两个表情图片为A2和A3。
步骤220，依据用户指令对所述展示信息进行显示处理。
对展示信息以弹泡的形式进行展示后，用户可以选择对候选项、展示信息等内容进行显示等，因此可以接收用户指令，然后按照用户指令对展示信息进行显示处理。其中，展示信息包括多种不同的信息，不同信息的显示方式也存在一定的差异，下面详细论述。
当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏。即可以依据用户指令对弹泡形式展示的表情图片进行选择，然后对选择出的表情图片进行上屏处理，即在当前输入环境的信息输入区域内显示该表情图片，其中表情图片可以与其对应候选项共同上屏，上屏时显示在同一信息框或不同显示框内，或者仅对该表情图片进行上屏，对候选项可以依据后续指示再进行处理。
当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对选择的多媒体信息进行播放，可以开启相应的播放器对该多媒体信息进行播放，也可以依据当前的应用程序的环境信息确定如何对多媒体信息进行显示处理，如在该应用程序中直接播放，又如对于即时通信应用还可以将该多媒体信息发送给好友(信息接收方)，推荐给好友查看。
当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页，可以调用浏览器等应用程序开启该网页链接信息对应的网页，还可以依据当前的应用程序的环境信息确定如何对网页链接信息进行显示处理，如对于即时通信应用还可以将该网页链接信息发送给好友，推荐给好友查看。
综上所述，对输入字符串对应的候选项结合上下文信息进行识别，可以识别出符合用户输入意图的展示信息。并且基于预置规则中的多种排序因素 对展示信息进行排序，获取到用户最可能选择的展示信息。
区别于在首屏候选项中展示表情图片，本实施例将展示信息展示在云推荐位置上，可以避免占用首屏候选项的位置，减少用户查找推荐信息和候选项的操作步骤，提高输入效率。
需要说明的是，对于方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明实施例并不受所描述的动作顺序的限制，因为依据本发明实施例，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作并不一定是本发明实施例所必须的。
实施例三
在上述实施例的基础上，本实施例还提供了一种输入法***。
参照图3，示出了本发明一种输入法***实施例的结构框图，具体可以包括如下模块：
关键词识别模块302，用于采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词。
展示信息匹配模块304，用于确定与所述语义关键词匹配的展示信息，其中，所述展示信息包括但不限于表情图片。
弹泡展示模块306，用于以弹泡形式对所述展示信息进行展示。
显示处理模块308，用于依据用户指令对所述展示信息进行显示处理，其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。
综上所述，采用上下文信息对接收的输入字符串进行识别处理，因此识别出的当前的语义关键词不限于输入字符串对应的候选项，而基于上下文信息的语义识别更加符合用户的输入意图。然后确定与所述语义关键词匹配的展示信息，因此该展示信息也更加符合用户的输入意图，以弹泡形式对所述展示信息进行展示，再依据用户指令对所述展示信息进行显示处理，如展示信息包括表情图片，则可以依据指令信息对表情图片进行上屏，从而可以在文字未上屏前就获取展示信息并展示，提高文字和表情等展示信息的输入效率。
参照图4，示出了本发明一种输入法***可选实施例的结构框图，具体可以包括如下模块：
所述关键词识别模块302，用于对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项；分别将所述上下文信息和每个候选项组合成候选语句；对每个候选语句分别进行语义分析，识别出当前的语义关键词。
所述展示信息匹配模块304，用于将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或，上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
该输入法***还包括：
更新模块310，用于检测服务器侧是否存在所述展示信息的更新数据包；并在存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地。
环境识别模块312，用于对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
排序模块314，用于按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序，其中所述预置规则中的排序因素包括但不限于：所述展示信息的使用次数、上次使用所述展示信息的时间、所述展示信息的推荐指数和用户的兴趣度。
所述弹泡展示模块306，用于按照所述排序顺序获取前N个展示信息；在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
所述展示信息还包括：多媒体信息和网页链接信息。
所述显示处理模块308，用于当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏；当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对选择的多媒体信息进行播放；当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页。
综上所述，对输入字符串对应的候选项结合上下文信息进行识别，可以识别出符合用户输入意图的展示信息。并且基于预置规则中的多种排序因素对展示信息进行排序，获取到用户最可能选择的展示信息。
区别于在首屏候选项中展示表情图片，本实施例将展示信息展示在云推荐位置上，可以避免占用首屏候选项的位置，减少用户查找推荐信息和候选项的操作步骤，提高输入效率。
对于装置实施例而言，由于其与方法实施例基本相似，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。
本领域内的技术人员应明白，本发明实施例的实施例可提供为方法、装置、或计算机程序产品。因此，本发明实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明实施例是参照根据本发明实施例的方法、终端设备(***)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理终端设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理终端设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理终端设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个 流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理终端设备上，使得在计算机或其他可编程终端设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程终端设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本发明实施例的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例做出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本发明实施例范围的所有变更和修改。
例如，图5示出了可以实现根据本发明的输入法虚拟键盘的调整的计算设备。该计算设备传统上包括处理器510和以存储器520形式的程序产品或者可读介质。存储器520可以是诸如闪存、EEPROM(电可擦除可编程只读存储器)、EPROM或者ROM之类的电子存储器。存储器520具有用于执行上述方法中的任何方法步骤的程序代码531的存储空间530。例如，用于程序代码的存储空间530可以包括分别用于实现上面的方法中的各种步骤的各个程序代码531。这些程序代码可以从一个或者多个程序产品中读出或者写入到这一个或者多个程序产品中。这些程序产品包括诸如存储卡之类的程序代码载体。这样的程序产品通常为如参考图6所述的便携式或者固定存储单元。该存储单元可以具有与图5的计算设备中的存储器520类似布置的存储段、存储空间等。程序代码可以例如以适当形式进行压缩。通常，存储单元包括可读代码531’，即可以由例如诸如510之类的处理器读取的代码，这些代码当由计算设备运行时，导致该计算设备执行上面所描述的方法中的各个步骤。
最后，还需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得 包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下，由语句“包括一个......”限定的要素，并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的相同要素。
以上对本发明所提供的一种基于输入的信息展示方法和一种输入法***，进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本发明的限制。

Claims (20)

  1. 一种基于输入的信息展示方法，其特征在于，包括：
    采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词；
    确定与所述语义关键词匹配的展示信息，并以弹泡形式对所述展示信息进行展示，其中，所述展示信息包括但不限于表情图片；
    依据用户指令对所述展示信息进行显示处理，其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。
  2. 根据权利要求1所述的方法，其特征在于，所述采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词的步骤，包括：
    对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项；
    分别将所述上下文信息和每个候选项组合成候选语句；
    对每个候选语句分别进行语义分析，识别出当前的语义关键词。
  3. 根据权利要求1所述的方法，其特征在于，所述确定与所述语义关键词匹配的展示信息，包括：
    将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或
    上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
  4. 根据权利要求1或3所述的方法，其特征在于，所述确定与所述语义关键词匹配的展示信息的步骤之前，还包括：
    检测服务器侧是否存在所述展示信息的更新数据包；
    当存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地。
  5. 根据权利要求4所述的方法，其特征在于，确定与所述语义关键词匹配的展示信息的步骤之后，还包括：
    对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
  6. 根据权利要求5所述的方法，其特征在于，还包括：
    按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序，其中所述预置规则中的排序因素包括但不限于：所述展示信息的使用次数、上次使用所述展示信息的时间、所述展示信息的推荐指数和用户的兴趣度。
  7. 根据权利要求6所述的方法，其特征在于，所述以弹泡形式对所述展示信息进行展示的步骤，包括：
    按照所述排序顺序获取前N个展示信息；
    在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
  8. 根据权利要求1所述的方法，其特征在于，所述展示信息还包括：多媒体信息和网页链接信息。
  9. 根据权利要求8所述的方法，其特征在于，所述依据用户指令对所述展示信息进行显示处理的步骤，包括：
    当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏；
    当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对选择的多媒体信息进行播放；
    当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页。
  10. 一种输入法***，其特征在于，包括：
    关键词识别模块，用于采用上下文信息对接收的输入字符串进行识别处理，识别出当前的语义关键词；
    展示信息匹配模块，用于确定与所述语义关键词匹配的展示信息，其中，所述展示信息包括但不限于表情图片；
    弹泡展示模块，用于以弹泡形式对所述展示信息进行展示；
    显示处理模块，用于依据用户指令对所述展示信息进行显示处理，其中，所述显示处理包括但不限于：对所述表情图片进行上屏操作。
  11. 根据权利要求10所述的***，其特征在于：
    所述关键词识别模块，用于对所述输入字符串进行转换，识别出所述输入字符串对应的各候选项；分别将所述上下文信息和每个候选项组合成候选语句；对每个候选语句分别进行语义分析，识别出当前的语义关键词。
  12. 根据权利要求10所述的***，其特征在于：
    所述展示信息匹配模块，用于将本地存储的展示信息与所述语义关键词进行匹配，确定与所述语义关键词匹配的展示信息；和/或，上传所述语义关键词，并接收服务器反馈的通过匹配确定的与所述语义关键词匹配的展示信息。
  13. 根据权利要求10或12所述的***，其特征在于，还包括：
    更新模块，用于检测服务器侧是否存在所述展示信息的更新数据包；并在存在所述展示信息的更新数据包时，下载所述更新数据包并存储到本地。
  14. 根据权利要求13所述的***，其特征在于，还包括：
    环境识别模块，用于对当前的环境信息进行识别，依据识别出的环境信息对所述与语义关键词匹配的展示信息进行筛选。
  15. 根据权利要求14所述的***，其特征在于，还包括：
    排序模块，用于按照预置规则对所述展示信息进行排序，确定所述展示信息的排序顺序，其中所述预置规则中的排序因素包括但不限于：所述展示信息的使用次数、上次使用所述展示信息的时间、所述展示信息的推荐指数和用户的兴趣度。
  16. 根据权利要求15所述的***，其特征在于：
    所述弹泡展示模块，用于按照所述排序顺序获取前N个展示信息；在当前输入法的云推荐位置以弹泡形式对所述前N个展示信息进行展示。
  17. 根据权利要求10所述的***，其特征在于，所述展示信息还包括：多媒体信息和网页链接信息。
  18. 根据权利要求17所述的***，其特征在于：
    所述显示处理模块，用于当所述展示信息为表情图片时，依据所述用户指令对所述表情图片进行选择，并对选择的表情图片进行上屏；当所述展示信息为多媒体信息时，依据所述用户指令对所述多媒体信息进行选择，并对 选择的多媒体信息进行播放；当所述展示信息为网页链接信息时，依据所述用户指令对所述网页链接信息进行选择，并开启选择网页链接信息对应的网页。
  19. 一种程序，包括可读代码，当所述可读代码在计算设备上运行时，导致所述计算设备执行根据权利要求1-9中的任一个所述的基于输入的信息展示方法。
  20. 一种可读介质，其中存储了如权利要求19所述的程序。
PCT/CN2015/087058 2014-09-25 2015-08-14 一种基于输入的信息展示方法和输入法*** WO2016045465A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/508,721 US10209782B2 (en) 2014-09-25 2015-08-14 Input-based information display method and input system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410499401.6 2014-09-25
CN201410499401.6A CN104298429B (zh) 2014-09-25 2014-09-25 一种基于输入的信息展示方法和输入法***

Publications (1)

Publication Number Publication Date
WO2016045465A1 true WO2016045465A1 (zh) 2016-03-31

Family

ID=52318180

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/087058 WO2016045465A1 (zh) 2014-09-25 2015-08-14 一种基于输入的信息展示方法和输入法***

Country Status (3)

Country Link
US (1) US10209782B2 (zh)
CN (1) CN104298429B (zh)
WO (1) WO2016045465A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108628911A (zh) * 2017-03-24 2018-10-09 微软技术许可有限责任公司 针对用户输入的表情预测
CN109814730A (zh) * 2017-11-20 2019-05-28 北京搜狗科技发展有限公司 输入方法和装置、用于输入的装置
CN110535669A (zh) * 2018-05-24 2019-12-03 连株式会社 内容推荐方法与内容推荐装置

Families Citing this family (63)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104298429B (zh) * 2014-09-25 2018-05-04 北京搜狗科技发展有限公司 一种基于输入的信息展示方法和输入法***
CN105988992B (zh) * 2015-03-02 2020-02-07 腾讯科技（深圳）有限公司 图标推送方法及装置
CN104866116A (zh) * 2015-03-25 2015-08-26 百度在线网络技术（北京）有限公司 一种用于输出表情信息的方法和装置
CN104809102B (zh) * 2015-04-01 2018-10-16 北京奇虎科技有限公司 一种基于输入的显示候选词的方法和装置
CN104881146A (zh) * 2015-04-28 2015-09-02 北京美秒科技有限公司 图片输入方法
CN106202087A (zh) * 2015-05-04 2016-12-07 阿里巴巴集团控股有限公司 一种信息推荐方法及装置
US9852131B2 (en) * 2015-05-18 2017-12-26 Google Llc Techniques for providing visual translation cards including contextually relevant definitions and examples
CN104917669A (zh) * 2015-05-29 2015-09-16 北京搜狗科技发展有限公司 一种信息处理方法和装置
CN104951099B (zh) * 2015-06-16 2017-12-19 北京奇虎科技有限公司 一种基于输入法的展示候选项的方法和装置
CN104933170B (zh) * 2015-06-30 2018-09-04 小米科技有限责任公司 信息展示方法及装置
CN105373236B (zh) * 2015-10-09 2020-09-08 北京搜狗科技发展有限公司 一种学词方法和装置
CN105425978A (zh) * 2015-10-26 2016-03-23 百度在线网络技术（北京）有限公司 输入数据的处理方法及装置
CN105549756A (zh) * 2015-10-30 2016-05-04 东莞酷派软件技术有限公司 一种基于位置信息的输入方法及用户终端
CN106648130A (zh) * 2015-11-02 2017-05-10 天脉聚源（北京）科技有限公司 一种表情符号输入方法及***
CN106789543A (zh) * 2015-11-20 2017-05-31 腾讯科技（深圳）有限公司 会话中实现表情图像发送的方法和装置
CN105404401A (zh) 2015-11-23 2016-03-16 小米科技有限责任公司 输入处理方法、装置及设备
CN105617657B (zh) * 2015-12-22 2019-06-18 北京奇虎科技有限公司 游戏智能推荐方法及装置
CN106257378A (zh) * 2016-07-19 2016-12-28 北京新美互通科技有限公司 一种表情符号输入方法及装置
CN105607756B (zh) * 2015-12-24 2019-05-10 百度在线网络技术（北京）有限公司 信息推荐方法和装置
CN107423277A (zh) * 2016-02-16 2017-12-01 中兴通讯股份有限公司 一种表情输入方法、装置及终端
CN105700703A (zh) * 2016-02-24 2016-06-22 北京小牛互联科技有限公司 一种在键盘的文字输入界面嵌入表情并支持自定义表情的方法和装置
CN107315487B (zh) * 2016-04-27 2022-04-12 北京搜狗科技发展有限公司 一种输入处理方法、装置及电子设备
CN106021279B (zh) * 2016-04-28 2020-04-07 百度在线网络技术（北京）有限公司 信息显示方法和装置
CN105929976A (zh) * 2016-05-25 2016-09-07 广州市久邦数码科技有限公司 一种基于输入法的动态表情输入方法及其***
CN106293120B (zh) * 2016-07-29 2020-06-23 维沃移动通信有限公司 表情输入方法及移动终端
CN106339159A (zh) * 2016-08-25 2017-01-18 乐视控股（北京）有限公司 字符选择处理方法、终端和语义分析服务器
CN106372059B (zh) * 2016-08-30 2018-09-11 北京百度网讯科技有限公司 信息输入方法和装置
CN106371624B (zh) * 2016-09-23 2019-03-19 百度在线网络技术（北京）有限公司 一种用于提供输入候选项的方法、装置和输入设备
CN106484139B (zh) * 2016-10-19 2019-01-29 北京新美互通科技有限公司 表情符号推荐方法及装置
WO2018119815A1 (zh) * 2016-12-29 2018-07-05 深圳前海达闼云端智能科技有限公司 图像显示方法、装置及电子设备
KR101804139B1 (ko) * 2017-02-15 2017-12-05 김진원 키워드 기반 데이터 관리 시스템 및 방법
CN108693979A (zh) * 2017-04-11 2018-10-23 北京搜狗科技发展有限公司 一种提供图片的方法、装置和设备
CN108803891B (zh) * 2017-04-28 2024-02-02 北京搜狗科技发展有限公司 一种信息展示方法、装置、电子设备以及存储介质
CN108874170B (zh) * 2017-05-08 2023-05-23 北京搜狗科技发展有限公司 一种输入方法及装置
CN109213332B (zh) * 2017-06-29 2022-11-08 北京搜狗科技发展有限公司 一种表情图片的输入方法和装置
CN110019883A (zh) * 2017-07-18 2019-07-16 腾讯科技（深圳）有限公司 获取表情图片的方法及装置
CN110019885B (zh) * 2017-08-01 2021-10-15 北京搜狗科技发展有限公司 一种表情数据推荐方法及装置
CN109388250B (zh) * 2017-08-04 2022-11-22 北京搜狗科技发展有限公司 一种信息推荐方法、装置及电子设备
CN107943382B (zh) * 2017-10-27 2020-06-05 贝壳找房(北京)科技有限公司 房源信息一键录入方法
CN109753205B (zh) * 2017-11-01 2021-02-02 北京搜狗科技发展有限公司 一种显示方法及装置
CN109799916B (zh) * 2017-11-16 2021-08-13 北京搜狗科技发展有限公司 一种候选项联想方法和装置
CN109947264B (zh) * 2017-12-21 2023-03-14 北京搜狗科技发展有限公司 一种信息展现方法、装置及电子设备
CN110110078B (zh) * 2018-01-11 2024-04-30 北京搜狗科技发展有限公司 数据处理方法和装置、用于数据处理的装置
CN108345391B (zh) * 2018-01-22 2020-06-26 平安科技（深圳）有限公司 文字排序方法、文字输入方法及终端设备
CN110096163A (zh) * 2018-01-29 2019-08-06 北京搜狗科技发展有限公司 一种表情输入方法及装置
CN110377766A (zh) * 2018-04-11 2019-10-25 北京搜狗科技发展有限公司 一种数据处理方法、装置和电子设备
CN108549637A (zh) * 2018-04-19 2018-09-18 京东方科技集团股份有限公司 基于拼音的语义识别方法、装置以及人机对话***
CN108595438A (zh) * 2018-05-08 2018-09-28 北京金山安全软件有限公司 信息处理方法、装置及设备
CN109036403A (zh) * 2018-07-20 2018-12-18 北京小米移动软件有限公司 信息交互方法及装置
CN109634436B (zh) * 2018-10-25 2023-11-10 平安科技（深圳）有限公司 输入法的联想方法、装置、设备及可读存储介质
CN109818847A (zh) * 2018-12-03 2019-05-28 北京奇安信科技有限公司 一种群聊表情的发送处理方法及装置
CN111435270B (zh) * 2019-01-11 2024-03-01 北京搜狗科技发展有限公司 一种推荐方法、装置和电子设备
CN109873756B (zh) * 2019-03-08 2020-04-03 百度在线网络技术（北京）有限公司 用于发送信息的方法和装置
CN110221710A (zh) * 2019-05-29 2019-09-10 北京金山安全软件有限公司 一种键盘输入方法、装置、电子设备及存储介质
CN110334144B (zh) * 2019-06-13 2023-07-25 平安银行股份有限公司 报文展示方法、装置、计算机设备及存储介质
CN110442248A (zh) * 2019-06-20 2019-11-12 上海萌家网络科技有限公司 一种输入法及输入***
CN110309517B (zh) * 2019-07-02 2023-10-17 北京百度网讯科技有限公司 表情文案处理方法、装置、***及存储介质
CN110505143A (zh) * 2019-08-07 2019-11-26 上海掌门科技有限公司 一种用于发送目标视频的方法与设备
CN111090762A (zh) * 2019-12-19 2020-05-01 京东方科技集团股份有限公司 图像获取方法、装置、电子设备及存储介质
CN113589953A (zh) * 2020-04-30 2021-11-02 北京搜狗科技发展有限公司 一种信息展示方法、装置和电子设备
CN113448461A (zh) * 2020-06-24 2021-09-28 北京新氧科技有限公司 信息处理方法、装置及设备
CN112035032B (zh) * 2020-06-30 2022-07-12 维沃移动通信有限公司 表情添加方法及装置
CN114816599B (zh) * 2021-01-22 2024-02-27 北京字跳网络技术有限公司 图像显示方法、装置、设备及介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130041911A1 (en) * 2011-08-10 2013-02-14 Nhn Corporation Method and apparatus for providing search services interactively displaying a type of search target
CN103809766A (zh) * 2012-11-06 2014-05-21 夏普株式会社 将文字转换成表情符号的方法和电子装置
CN103853446A (zh) * 2012-11-29 2014-06-11 北京千橡网景科技发展有限公司 用于智能输入的方法和设备
CN104298429A (zh) * 2014-09-25 2015-01-21 北京搜狗科技发展有限公司 一种基于输入的信息展示方法和输入法***

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7539699B2 (en) * 2003-03-21 2009-05-26 Yahoo! Inc. Apparatus, system and method for use in generating and maintaining an electronic address book
JP2005135169A (ja) * 2003-10-30 2005-05-26 Nec Corp 携帯端末およびデータ処理方法
JP4341408B2 (ja) * 2004-01-15 2009-10-07 パナソニック株式会社 画像表示方法及び装置
WO2007069573A1 (ja) * 2005-12-16 2007-06-21 Matsushita Electric Industrial Co., Ltd. 移動体用入力装置、及び方法
KR101138396B1 (ko) * 2007-09-11 2012-04-26 삼성전자주식회사 Ｉｐｔｖ 단말기에서 컨텐츠 재생 방법 및 장치
US8060471B2 (en) * 2008-12-11 2011-11-15 Inventec Besta Co., Ltd. Updating display system of display list and method thereof
CN101593073A (zh) * 2009-07-10 2009-12-02 张向阳 一种集成在输入法中的广告呈现设备
CN104063427A (zh) * 2014-06-06 2014-09-24 北京搜狗科技发展有限公司 一种基于语义理解的表情输入方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130041911A1 (en) * 2011-08-10 2013-02-14 Nhn Corporation Method and apparatus for providing search services interactively displaying a type of search target
CN103809766A (zh) * 2012-11-06 2014-05-21 夏普株式会社 将文字转换成表情符号的方法和电子装置
CN103853446A (zh) * 2012-11-29 2014-06-11 北京千橡网景科技发展有限公司 用于智能输入的方法和设备
CN104298429A (zh) * 2014-09-25 2015-01-21 北京搜狗科技发展有限公司 一种基于输入的信息展示方法和输入法***

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108628911A (zh) * 2017-03-24 2018-10-09 微软技术许可有限责任公司 针对用户输入的表情预测
CN108628911B (zh) * 2017-03-24 2022-09-06 微软技术许可有限责任公司 针对用户输入的表情预测
CN109814730A (zh) * 2017-11-20 2019-05-28 北京搜狗科技发展有限公司 输入方法和装置、用于输入的装置
CN109814730B (zh) * 2017-11-20 2023-09-12 北京搜狗科技发展有限公司 输入方法和装置、用于输入的装置
CN110535669A (zh) * 2018-05-24 2019-12-03 连株式会社 内容推荐方法与内容推荐装置
CN110535669B (zh) * 2018-05-24 2022-11-25 连株式会社 内容推荐方法与内容推荐装置

Also Published As

Publication number Publication date
US10209782B2 (en) 2019-02-19
CN104298429B (zh) 2018-05-04
US20170277277A1 (en) 2017-09-28
CN104298429A (zh) 2015-01-21

Similar Documents

Publication Publication Date Title
WO2016045465A1 (zh) 一种基于输入的信息展示方法和输入法***
US11308173B2 (en) Searching for ideograms in an online social network
US11797622B2 (en) Text-to-media indexes on online social networks
US11315546B2 (en) Computerized system and method for formatted transcription of multimedia content
US10402703B2 (en) Training image-recognition systems using a joint embedding model on online social networks
US11138207B2 (en) Integrated dynamic interface for expression-based retrieval of expressive media content
US10650009B2 (en) Generating news headlines on online social networks
US10685186B2 (en) Semantic understanding based emoji input method and device
US20180101540A1 (en) Diversifying Media Search Results on Online Social Networks
EP2795441B1 (en) Systems and methods for identifying and suggesting emoticons
US10083379B2 (en) Training image-recognition systems based on search queries on online social networks
US20170154104A1 (en) Real-time recommendation of reference documents
JP6776310B2 (ja) ユーザ−入力コンテンツと連関するリアルタイムフィードバック情報提供方法およびシステム
US20220092071A1 (en) Integrated Dynamic Interface for Expression-Based Retrieval of Expressive Media Content
CN107924398B (zh) 用于提供以评论为中心的新闻阅读器的***和方法
US20230353820A1 (en) Systems and Methods for Improved Searching and Categorizing of Media Content Items Based on a Destination for the Media Content Machine Learning
CN111222316B (zh) 文本检测方法、装置及存储介质
CN106663123B (zh) 以评论为中心的新闻阅读器
CN111767259A (zh) 内容分享的方法、装置、可读介质和电子设备
EP3306555A1 (en) Diversifying media search results on online social networks
CN110929122A (zh) 一种数据处理方法、装置和用于数据处理的装置
CN113407040A (zh) 一种信息处理方法、装置和介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15843998

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15508721

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15843998

Country of ref document: EP

Kind code of ref document: A1