CN112559699A - Information interaction method, device and equipment - Google Patents
Information interaction method, device and equipment Download PDFInfo
- Publication number
- CN112559699A CN112559699A CN202011237542.2A CN202011237542A CN112559699A CN 112559699 A CN112559699 A CN 112559699A CN 202011237542 A CN202011237542 A CN 202011237542A CN 112559699 A CN112559699 A CN 112559699A
- Authority
- CN
- China
- Prior art keywords
- word
- information
- processed
- sentence
- intention
- 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.)
- Pending
Links
Images
Classifications
-
- 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
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses an information interaction method, equipment and equipment, wherein the method comprises the following steps: receiving information to be processed; detecting whether a new word exists in the information to be processed or not, searching a reference word and sentence which contains the new word and is matched with the information to be processed from an intention combination library when a detection result shows that the new word exists in the information to be processed, and acquiring a first answer intention of a first answer word and sentence aiming at the reference word and sentence; and outputting response information aiming at the information to be processed according to the first response intention. In this way, when a new word is found, a response is made based on similar words in the intention combination library, and the intention combination library may be a combination of a reference word sentence containing the new word and an intention of a response word sentence or an answer word sentence determined based on intention recognition of the following content of a word sentence containing the new word when the new word is found up to then. Therefore, the tedious process of collecting the corpus updating model is effectively avoided, words and sentences containing new words are quickly responded, and the labor conversion rate is obviously reduced.
Description
Technical Field
The invention relates to the technical field of natural language processing, in particular to an information interaction method, device and equipment.
Background
With the development of networks, the information propagation speed is exponentially increased, so that some new words or new word combination modes have contextual changes in meanings as the times change, so that the dialog system cannot correctly understand the user intention, for example: eating individual melon, wool, etc. And these new words may be popular only for a while, with a short life cycle. The stage characteristic of the method can greatly influence the understanding accuracy of a bot (search engine grabbing robot, also called a crawler), and after the corpora are artificially collected and the model is updated, the life cycle of the corresponding word is possibly finished, and the meaning is rarely or rarely changed when the word appears again.
In contrast, the current solution is to find new words based on rules or statistical methods, and manually interpret the new words or find similar words for replacement. However, if only the new word text is found, no contribution is made to the understanding of the new word, and the problem of understanding and responding to the user information still cannot be solved.
Disclosure of Invention
In order to solve the above problems, embodiments of the present invention creatively provide an information interaction method, apparatus, and device.
According to a first aspect of the present invention, there is provided an information interaction method, including: receiving information to be processed; detecting whether new words exist in the information to be processed or not to obtain a new word detection result, wherein the new words comprise words formed by specific word combinations and words with meanings suitable for specific language environments in the information to be processed; when the new word detection result shows that a new word exists in the information to be processed, searching reference words and sentences which contain the new word and are matched with the information to be processed from an intention combination library to obtain a search result; acquiring a first response intention of a first response word sentence aiming at the reference word sentence according to the search result; and outputting response information aiming at the information to be processed according to the first response intention.
According to an embodiment of the invention, the method further comprises: when the search result shows that the intention combination library does not have the reference word or sentence matched with the information to be processed, receiving a second response word or sentence aiming at the information to be processed; identifying a second response intent of the second response word; detecting and recording the accumulated times of receiving the information to be processed containing the new words and corresponding second response words and sentences with the same second response intention within a first set time; and when the accumulated times reach a first set threshold value, updating the intention combination library according to the information to be processed, the second answer words and sentences and the second answer intention.
According to an embodiment of the present invention, the intent composition library includes: a plurality of reference sentences containing new words; and a first answer word and/or a first answer intention for each of said reference sentences.
According to an embodiment of the present invention, the intent composition library further comprises: and aiming at the set validity period of each reference sentence, wherein the set validity period is used for deleting the reference sentence or marking the reference sentence as an invalid sentence when the existence duration of the reference sentence in the intention combination library exceeds the set validity period.
According to an embodiment of the invention, the method further comprises: according to a set period, determining the matching times of each reference word and sentence in the intention combination library according to the search result; and adjusting the set validity period of the corresponding reference words and sentences according to the matching times.
According to an embodiment of the present invention, the adjusting the set validity period of the corresponding reference word according to the matching times includes: for each reference word, reducing the set validity period of the corresponding reference word when the matching times satisfy at least one of the following: the first matching times determined in the latest set period are smaller than the second matching times determined in the previous set period, and the difference between the first matching times and the second matching times is smaller than a first set difference; the first number of matches is less than a first match threshold.
According to an embodiment of the present invention, the adjusting the set validity period of the corresponding reference word according to the matching times includes: for each reference word, when the matching times satisfy at least one of the following, increasing the set validity period of the corresponding reference word: the first matching times determined in the latest set period are greater than the second matching times determined in the previous set period, and the absolute value of the difference between the first matching times and the second matching times is greater than a second set difference; the first number of matches is greater than a second match threshold.
According to an embodiment of the present invention, the adjusting the set validity period of the corresponding reference word according to the matching times includes: and when the first matching times determined in the latest set period is less than a third matching threshold, adjusting the set validity period of the corresponding reference sentence to zero.
According to the second aspect of the present invention, there is also provided an information interaction apparatus, the apparatus including: the information receiving module is used for receiving information to be processed; the new word detection module is used for detecting whether a new word exists in the information to be processed to obtain a new word detection result, wherein the new word comprises a word formed by specific word combination and a word with a meaning suitable for a specific language environment in the information to be processed; the new word matching module is used for searching reference words and sentences which contain the new words and are matched with the information to be processed from an intention combination library to obtain a search result when the new word detection result shows that the new words exist in the information to be processed; the intention understanding module is used for acquiring a first response intention of a first response word sentence aiming at the reference word sentence according to the search result; and the response module is used for outputting response information aiming at the information to be processed according to the first response intention.
According to a third aspect of the present invention, there is also provided an apparatus comprising at least one processor, and at least one memory connected to the processor, a bus; the processor and the memory complete mutual communication through the bus; the processor is used for calling the program instructions in the memory so as to execute the information interaction method.
According to a fourth aspect of the present invention, there is also provided a computer-readable storage medium comprising a set of computer-executable instructions, which when executed, are configured to perform any of the above-mentioned information interaction methods.
The information interaction method, the equipment and the equipment in the embodiment of the invention receive information to be processed; detecting whether a new word exists in the information to be processed or not, searching a reference word and sentence which contains the new word and is matched with the information to be processed from an intention combination library when a detection result shows that the new word exists in the information to be processed, and acquiring a first answer intention of a first answer word and sentence aiming at the reference word and sentence; and outputting response information aiming at the information to be processed according to the first response intention. In this way, when a new word is found, a response is made based on similar words in the intention combination library, and the intention combination library may be a combination of a reference word sentence containing the new word and response information determined based on intention recognition of the following content of a word sentence containing the new word when the new word is found up to then, or a combination of a word sentence containing the new word and an intention of a response word sentence. No matter the meaning of the words is transferred or new words or new word sequences appear, the dynamic updating of the intention combination library can support the dialogue system to quickly respond to the understanding of the intention words with short life cycles, effectively avoid the tedious process of collecting a corpus updating model, quickly respond to the words and sentences containing the new words, and obviously reduce the labor conversion rate.
It is to be understood that the teachings of the present invention need not achieve all of the above-described benefits, but rather that specific embodiments may achieve specific technical results, and that other embodiments of the present invention may achieve benefits not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a schematic diagram illustrating an implementation flow of an information interaction method according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating an implementation of an example of an application of the information interaction method according to the embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a structure of an information interaction apparatus according to an embodiment of the present invention;
fig. 4 is a schematic diagram showing a composition structure of the apparatus according to the embodiment of the present invention.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given only to enable those skilled in the art to better understand and to implement the present invention, and do not limit the scope of the present invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The technical solution of the present invention is further elaborated below with reference to the drawings and the specific embodiments.
The embodiment of the invention can be applied to scenes with an information interaction function, such as intelligent customer service, and the like, and responds to the information input by the user according to the voice information or the character information and the like input by the user. Of course, the present invention may also be applied to other suitable information interaction scenarios, and the present invention is not limited thereto.
Fig. 1 shows a schematic flow chart of an implementation of the information interaction method according to the embodiment of the present invention.
Referring to fig. 1, an information interaction method according to an embodiment of the present invention at least includes the following operation flows: operation 101, receiving information to be processed; operation 102, detecting whether a new word exists in the information to be processed to obtain a new word detection result, where the new word includes a word formed by a specific word combination and a word having a meaning suitable for a specific language environment in the information to be processed; operation 103, when the new word detection result shows that a new word exists in the information to be processed, searching a reference word and sentence which contains the new word and is matched with the information to be processed from the intention combination library to obtain a search result; operation 104, obtaining a first response intention of a first response word sentence for the reference word sentence according to the search result; in operation 105, response information for the information to be processed is output according to the first response intention.
In operation 101, information to be processed is received.
The information to be processed may be voice information, text information, etc. For example: for intelligent customer service, chat robots and the like, receiving character information sent by users and responding to information to be processed is needed. The information to be processed can also be information to be processed, such as an intelligent assistant of a chat group in the instant chat tool.
And operation 102, detecting whether a new word exists in the information to be processed to obtain a new word detection result, wherein the new word comprises a word formed by a specific word combination and a word having a meaning suitable for a specific language environment in the information to be processed.
In one embodiment of the present invention, the new word may be a completely new combination of words, such as: the sauce purple is originally two irrelevant Chinese characters which are combined together and have no specific meaning, but is actually understood as the sauce purple in network language. The new word may also be a word having a conventional meaning or a new meaning of a word, such as: "wool" itself refers to the long wool of sheep, but the description of "wool in pulling" appears in some online shopping communication, etc., which is understood as seizing the loophole of the merchant or client, and getting the commodity or purchase ticket with higher value at lower price.
Whether a new word exists in the information to be processed can be determined in one of the following ways, and other applicable methods can also be adopted:
1. and maintaining an interactive word list of information interaction, and if words which are not contained in the interactive word list are detected to appear in the currently received information to be processed, determining that the information to be processed contains new words.
2. Acquiring a large amount of information interaction resources, training a one-class svm (one-class support vector machine) model, and then performing new word detection on the received information to be processed by using the trained one-class svm.
3. Emotion detection techniques in natural language processing are employed, for example: and determining whether the bot correctly understands the received information to be processed sent by the user according to whether the user has negative emotion or not in the interaction process of the bot (crawler) and the user.
Of course, the embodiment of the present invention may also adopt other applicable new word detection methods, and is not limited to the above manner. And the number of the new words in the information to be processed can be one or more.
In operation 103, when the new word detection result shows that a new word exists in the information to be processed, searching a reference word and sentence which contains the new word and is matched with the information to be processed from the intention combination library to obtain a search result;
in one embodiment of the invention, the intent composition library is generated prior to receiving the current new word. Specifically, after the new word is recognized, the intention combination library may be generated by receiving the information interaction content between the server and the client, and performing operations such as intention recognition according to the received information interaction content, and the intention combination library may be stored in a form of a combination pair of < information to be processed including the new word, intention of interaction information in a set round >, or in another applicable manner.
For example, in an intelligent assistance scene in intelligent customer service or instant chat, it is determined that machines such as intelligent customer service cannot directly talk with a user, and at this time, manual work is switched over, information interaction contents of the manual customer and the user are received, and the intention of each conversation content in the set turn conversation contents after a new word appears is determined. And constructing an intention combination pair of the words and sentences containing the new words and the intention of the next interactive content, or constructing an intention combination pair of the words and sentences containing the new words and a plurality of intentions of the interactive content of the next set turn.
In one embodiment of the present invention, the intent composition library includes: a plurality of reference sentences containing new words; and a first answer word sentence and/or a first answer intention for each reference sentence.
In an embodiment of the present invention, the intent composition library further includes: and aiming at the set validity period of each reference sentence, the set validity period is used for deleting the reference sentence or marking the reference sentence as an invalid sentence when the existence duration of the reference sentence in the intention combination library exceeds the set validity period.
In operations 104 to 105, a first response intention of a first response word sentence with respect to the reference word sentence is obtained according to the search result, and response information with respect to the information to be processed is output according to the first response intention.
In an embodiment of the present invention, the first answer sentence may be a dialog of a set turn, and the first answer intention may be a combination of a plurality of intentions. Likewise, the second answer phrase may also be a dialog setting round, and the second answer intention may be a combination of a plurality of intentions, hereinafter.
For example, the intent combination library may include pairs of intent combinations generated from the following dialog content:
"A: give your individual melons.
A: girl friend and his deciduous hand in 9X stars.
B: i also hear and say cheerful, really do not want. "
The intent to identify the above is to say that the intent "female friend X star 9 and his hands are cheering" is entertainment news, and the intent "i also listen to say cheering, really do not want a". Thus, for the word sentence "send out a melon to your" containing the new word "melon", the intent of the following interactive information may be "entertainment news", and generate < send out a melon to your; entertainment news, discussion heat … … >.
Intent combination pairs in the intent combination library may also be generated from the following dialog content:
"A: does the group have wool from diapers recently?
B: new low price … … no … … diapers … … (commodity link) limited to 1000 new rolls @ handbook.
A: easy to calculate, purchased "
The intention of recognizing the above is to recognize that the intention of "new low … … limited 1000 new rolls @ manual no … … diaper … …" is a commodity link and a price advantage, "good cost, purchased" is an intention of ordering purchase. Thus, for the word "does the group have a recent diaper wool" that contains the new word "wool", the intent of the following interactive information may be "entertainment news", and generate < does wool with XX; the intent combination pairs of merchandise link, price offer, order purchase … … >.
If the received information to be processed is 'wool for skin cream', finding wool with XX from the intention combination library; the intent combination pairs of merchandise link, price offer, order purchase … … >. And according to the intention combination pair, determining that the response information of the information to be processed, namely 'seeking the wool of the skin cream', is a commodity link with price preference of the skin cream.
In an embodiment of the present invention, when the search result shows that the intention combination base does not have a reference word or phrase matching the information to be processed, a second answer word or phrase for the information to be processed is received; identifying a second response intent of the second response word; detecting and recording the accumulated times of receiving the information to be processed containing the new words within the first set time and the second response intentions of the corresponding second response words and sentences are the same; and when the accumulated times reach a first set threshold value, updating the intention combination library according to the information to be processed, the second answer words and sentences and the second answer intention.
In one embodiment of the invention, the matching times of each reference word and sentence in the intention combination library are determined according to the set period and the search result; and adjusting the set validity period of the corresponding reference words and sentences according to the matching times.
For example, the default set validity period may be 63 days when generating an intent combination pair. Setting a period of 7 days, confirming once every 7 days, and matching times of each reference word and sentence in the intention combination library within 7 days. For example: if the received information to be processed is 'wool for skin cream', the wool with XX can be searched from the intention combination library; the intent combination pairs of merchandise link, price offer, order purchase … … >. And according to the intention combination pair, determining that the response information of the information to be processed, namely 'seeking the wool of the skin cream', is a commodity link with price preference of the skin cream. Thus, the reference word "wool with XX" is considered to be matched, and the number of matching times of the reference word is increased by 1. And by analogy, according to a set period and according to a search result, determining the matching times of each reference word and sentence in the intention combination library. If the reference word or sentence is matched for a plurality of times, the reference word or sentence still has the intention recorded in the intention combination library, and the practical value is high. According to the rule, the set validity period of the corresponding reference words and sentences is adjusted, so that the timeliness of the intention combination library is effectively guaranteed, the reference words and sentences in the intention combination library are timely updated, meanwhile, the memory and the computing resources occupied by the intention combination library are accurately controlled, and the response speed of the intention combination library to the information to be processed is remarkably improved.
In one embodiment of the present invention, the following operation steps are adopted to adjust the set validity period of the corresponding reference word and sentence according to the matching times: and aiming at each reference word and sentence, when the matching times meet at least one of the following conditions, reducing the set validity period of the corresponding reference word and sentence: the first matching times determined in the latest setting period are smaller than the second matching times determined in the previous setting period, and the difference value between the first matching times and the second matching times is smaller than the first setting difference value; the first number of matches is less than a first match threshold.
The first matching threshold may be a predetermined fixed value, such as 100 times, 200 times, and the like. Or it may be a value that varies with other parameters, such as: and the mean value of the matching times of the reference words and sentences in all the setting periods before the current setting period. Similarly, the second matching threshold and the third matching threshold may be fixed values preset in the following, or may be values that vary with other parameters.
For example, for the reference word a, the initial validity period is set to 63 days, the period is set to 7 days, the number of matching times for the reference word a in the fifth 7 days is 2 times, the number of matching times for the reference word a in the fourth 7 days is 320 times, the first set difference value is 200 times, the average value of the number of matching times in the first four set periods is 300 times, and the threshold value is set to 180. Obviously, the first matching time 2 determined in the latest setting period is less than the second matching time 320 determined in the previous setting period, and the difference 318 between the first matching time and the second matching time is less than the first setting difference 200; the first number of matches 2 is less than the first match threshold 180. Therefore, the set validity period of the corresponding reference words is reduced. The reduction of the set validity period at most can be set according to practical situations, for example: the matching times can be detected each time, the validity period of the corresponding reference sentence is reduced by 7 days, and the like.
In an embodiment of the present invention, adjusting the set validity period of the corresponding reference word according to the matching times includes: aiming at each reference word and sentence, when the matching times meet at least one of the following conditions, the set validity period of the corresponding reference word and sentence is increased: the first matching times determined in the latest set period are greater than the second matching times determined in the previous set period, and the absolute value of the difference between the first matching times and the second matching times is greater than the second set difference; the first number of matches is greater than a second match threshold.
For example, for the reference word B, the initial validity period is set to 63 days, the period is set to 7 days, the number of times of matching with respect to the reference word B in the ninth 7 days is 600 times, the number of times of matching with respect to the reference word B in the eighth 7 days is 320 times, the second setting difference value is 200 times, the average value of the number of times of matching in the first four setting periods is 300 times, and the threshold value is set to 180. Obviously, the first matching times 600 determined in the latest setting period is greater than the second matching times 320 determined in the previous setting period, and the absolute value 280 of the difference between the first matching times 600 and the second matching times is greater than the second setting difference 200; the first number of matches 600 is greater than the second match threshold 320. Therefore, the set validity period of the corresponding reference sentence is increased. How much the set validity period is increased can be set according to actual conditions, such as: the validity period of the corresponding reference sentence is increased by 7 days and the like when the matching times are met every time.
In an embodiment of the present invention, adjusting the set validity period of the corresponding reference word according to the matching times includes: and when the first matching times determined in the latest setting period is less than the third matching threshold, adjusting the setting validity period of the corresponding reference word sentence to be zero.
For example, for the reference word C, the initial validity period is set to 63 days, the period is set to 7 days, the matching frequency for the reference word a in the ninth 7 days is 0, and the third matching threshold is 1. The set validity period of the reference word sentence C is adjusted to zero when the number of matches for the reference word sentence a detected at the end of the ninth 7 days is detected to be 0.
Fig. 2 is a schematic diagram illustrating an implementation flow of a specific application example of the information interaction method according to the embodiment of the present invention.
Referring to fig. 2, a specific application example of the information interaction method according to the embodiment of the present invention at least includes the following operation flows:
in operation 201, information to be processed is received.
At operation 203, similar statements are looked up from the intent composition library. If so, operation 203 is performed, otherwise, operation 207 is performed.
In operation 205, a chat log is received and intent analysis is performed according to the chat content of the set turn.
Specifically, after switching to manual operation, receiving manual operation of the server and interactive information of the user, for example: and (4) dialog information.
In operation 206, it is determined whether the frequency of occurrence of new words in the information to be processed exceeds a predetermined frequency. If so, an intention combination pair is generated according to the intention recognition performed in operation 205 and the new words included in the corresponding information to be processed, and is stored in an intention combination library. Therefore, when a new word is detected, the intention combination pair is not immediately obtained according to the analysis of the words and the sentences containing the new word and the following content thereof, the frequency of the new word is detected, and if the frequency of the new word exceeds a set threshold value, the corresponding intention combination of the new word and the following content thereof is stored in the intention combination library, so that the timeliness of the intention combination library is ensured, and the stability of the intention combination library is ensured.
At operation 208, interactions are performed according to the intent combination templates in the intent combination library.
The information interaction method, the equipment and the equipment in the embodiment of the invention receive information to be processed; detecting whether new words exist in the information to be processed or not, searching reference words and sentences which contain the new words and are matched with the information to be processed from an intention combination library when a detection result shows that the new words exist in the information to be processed, and acquiring a first answer intention of a first answer word and sentence aiming at the reference words and sentences; and outputting response information aiming at the information to be processed according to the first response intention. In this way, when a new word is found, a response is made based on similar words in the intention combination library, and the intention combination library may be a combination of a reference word sentence containing the new word and response information determined based on intention recognition of the following content of a word sentence containing the new word when the new word is found up to then, or a combination of a word sentence containing the new word and an intention of a response word sentence. No matter the meaning of the words is transferred or new words or new word sequences appear, the dynamic updating of the intention combination library can support the dialogue system to quickly respond to the understanding of the intention words with short life cycles, effectively avoid the tedious process of collecting a corpus updating model, quickly respond to the words and sentences containing the new words, and obviously reduce the labor conversion rate.
Similarly, based on the above information processing method, an embodiment of the present invention further provides a computer-readable storage medium, in which a program is stored, and when the program is executed by a processor, the processor is caused to perform at least the following operation steps: operation 101, receiving information to be processed; operation 102, detecting whether a new word exists in the information to be processed to obtain a new word detection result, where the new word includes a word formed by a specific word combination and a word having a meaning suitable for a specific language environment in the information to be processed; operation 103, when the new word detection result shows that a new word exists in the information to be processed, searching a reference word and sentence which contains the new word and is matched with the information to be processed from the intention combination library to obtain a search result; operation 104, obtaining a first response intention of a first response word sentence for the reference word sentence according to the search result; in operation 105, response information for the information to be processed is output according to the first response intention.
Further, based on the above information processing method, an embodiment of the present invention further provides an information interaction apparatus, as shown in fig. 3, where the apparatus 30 includes: an information receiving module 301, configured to receive information to be processed; a new word detection module 302, configured to detect whether a new word exists in the information to be processed, to obtain a new word detection result, where the new word includes a word formed by a specific word combination and a word having a meaning suitable for a specific language environment in the information to be processed; a new word matching module 303, configured to, when the new word detection result shows that a new word exists in the information to be processed, search for a reference word and sentence that includes the new word and matches the information to be processed from the intention combination library, and obtain a search result; an intention understanding module 304, configured to obtain a first response intention of a first response word sentence for the reference word sentence according to the search result; a response module 305, configured to output response information for the information to be processed according to the first response intention.
Still further, based on the information processing method described above, an embodiment of the present invention further provides an apparatus, and referring to fig. 4, the apparatus 40 includes at least one processor 401, and at least one memory 402 and a bus 403 connected to the processor 401; the processor 401 and the memory 402 complete communication with each other through the bus 403; the processor 401 is used for calling the program instructions in the memory 402 to execute the above information interaction method.
Here, it should be noted that: the above description of the embodiments of the information interaction apparatus and device is similar to the description of the method embodiment shown in fig. 1 to 2, and has similar beneficial effects to the method embodiment shown in fig. 1 to 2, and therefore, the description is omitted. For technical details that are not disclosed in the embodiments of the information interaction apparatus and device of the present invention, please refer to the description of the method embodiments shown in fig. 1 to 2 of the present invention for understanding, and therefore, for brevity, will not be described again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of a unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. An information interaction method, the method comprising:
receiving information to be processed;
detecting whether new words exist in the information to be processed or not to obtain a new word detection result, wherein the new words comprise words formed by specific word combinations and words with meanings suitable for specific language environments in the information to be processed;
when the new word detection result shows that a new word exists in the information to be processed, searching reference words and sentences which contain the new word and are matched with the information to be processed from an intention combination library to obtain a search result;
acquiring a first response intention of a first response word sentence aiming at the reference word sentence according to the search result;
and outputting response information aiming at the information to be processed according to the first response intention.
2. The method of claim 1, further comprising:
when the search result shows that the intention combination library does not have the reference word or sentence matched with the information to be processed, receiving a second response word or sentence aiming at the information to be processed;
identifying a second response intent of the second response word;
detecting and recording the accumulated times of receiving the information to be processed containing the new words and corresponding second response words and sentences with the same second response intention within a first set time;
and when the accumulated times reach a first set threshold value, updating the intention combination library according to the information to be processed, the second answer words and sentences and the second answer intention.
3. The method of claim 1, the library of intent combinations comprising:
a plurality of reference sentences containing new words; and
a first answer word sentence and/or a first answer intention for each of the reference sentences.
4. The method of claim 3, the intent composition library further comprising:
and aiming at the set validity period of each reference sentence, wherein the set validity period is used for deleting the reference sentence or marking the reference sentence as an invalid sentence when the existence duration of the reference sentence in the intention combination library exceeds the set validity period.
5. The method of claim 4, further comprising:
according to a set period, determining the matching times of each reference word and sentence in the intention combination library according to the search result;
and adjusting the set validity period of the corresponding reference words and sentences according to the matching times.
6. The method of claim 5, wherein said adjusting the set validity period of the corresponding reference sentence according to the number of matches comprises: for each reference word, reducing the set validity period of the corresponding reference word when the matching times satisfy at least one of the following:
the first matching times determined in the latest set period are smaller than the second matching times determined in the previous set period, and the difference between the first matching times and the second matching times is smaller than a first set difference;
the first number of matches is less than a first match threshold.
7. The method of claim 5, wherein said adjusting the set validity period of the corresponding reference sentence according to the number of matches comprises: for each reference word, when the matching times satisfy at least one of the following, increasing the set validity period of the corresponding reference word:
the first matching times determined in the latest set period are greater than the second matching times determined in the previous set period, and the absolute value of the difference between the first matching times and the second matching times is greater than a second set difference;
the first number of matches is greater than a second match threshold.
8. The method of claim 5, wherein said adjusting the set validity period of the corresponding reference sentence according to the number of matches comprises: and when the first matching times determined in the latest set period is less than a third matching threshold, adjusting the set validity period of the corresponding reference sentence to zero.
9. An information interaction apparatus, the apparatus comprising:
the information receiving module is used for receiving information to be processed;
the new word detection module is used for detecting whether a new word exists in the information to be processed to obtain a new word detection result, wherein the new word comprises a word formed by specific word combination and a word with a meaning suitable for a specific language environment in the information to be processed;
the new word matching module is used for searching reference words and sentences which contain the new words and are matched with the information to be processed from an intention combination library to obtain a search result when the new word detection result shows that the new words exist in the information to be processed;
the intention understanding module is used for acquiring a first response intention of a first response word sentence aiming at the reference word sentence according to the search result;
and the response module is used for outputting response information aiming at the information to be processed according to the first response intention.
10. A device comprising at least one processor, and at least one memory, bus connected with the processor; the processor and the memory complete mutual communication through the bus; the processor is configured to call program instructions in the memory to perform the information interaction method of any one of claims 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011237542.2A CN112559699A (en) | 2020-11-09 | 2020-11-09 | Information interaction method, device and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011237542.2A CN112559699A (en) | 2020-11-09 | 2020-11-09 | Information interaction method, device and equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112559699A true CN112559699A (en) | 2021-03-26 |
Family
ID=75041728
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011237542.2A Pending CN112559699A (en) | 2020-11-09 | 2020-11-09 | Information interaction method, device and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112559699A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102073527A (en) * | 2011-01-26 | 2011-05-25 | 百度在线网络技术(北京)有限公司 | Method and device for updating input method word stock |
CN102163198A (en) * | 2010-02-24 | 2011-08-24 | 北京搜狗科技发展有限公司 | A method and a system for providing new or popular terms |
CN109635091A (en) * | 2018-12-14 | 2019-04-16 | 上海钛米机器人科技有限公司 | A kind of method for recognizing semantics, device, terminal device and storage medium |
CN110489517A (en) * | 2018-05-09 | 2019-11-22 | 鼎捷软件股份有限公司 | The Auto-learning Method and system of virtual assistant |
CN111125327A (en) * | 2019-12-11 | 2020-05-08 | 中国建设银行股份有限公司 | Short-session-based new word discovery method, storage medium and electronic device |
-
2020
- 2020-11-09 CN CN202011237542.2A patent/CN112559699A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102163198A (en) * | 2010-02-24 | 2011-08-24 | 北京搜狗科技发展有限公司 | A method and a system for providing new or popular terms |
CN102073527A (en) * | 2011-01-26 | 2011-05-25 | 百度在线网络技术(北京)有限公司 | Method and device for updating input method word stock |
CN110489517A (en) * | 2018-05-09 | 2019-11-22 | 鼎捷软件股份有限公司 | The Auto-learning Method and system of virtual assistant |
CN109635091A (en) * | 2018-12-14 | 2019-04-16 | 上海钛米机器人科技有限公司 | A kind of method for recognizing semantics, device, terminal device and storage medium |
CN111125327A (en) * | 2019-12-11 | 2020-05-08 | 中国建设银行股份有限公司 | Short-session-based new word discovery method, storage medium and electronic device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108536852B (en) | Question-answer interaction method and device, computer equipment and computer readable storage medium | |
CN104951428B (en) | User's intension recognizing method and device | |
WO2016197767A2 (en) | Method and device for inputting expression, terminal, and computer readable storage medium | |
CN112487173B (en) | Man-machine conversation method, device and storage medium | |
CN110597952A (en) | Information processing method, server, and computer storage medium | |
CN110187780B (en) | Long text prediction method, long text prediction device, long text prediction equipment and storage medium | |
JP2023535709A (en) | Language expression model system, pre-training method, device, device and medium | |
WO2013134871A1 (en) | System and method for conversation-based information search | |
CN109284502B (en) | Text similarity calculation method and device, electronic equipment and storage medium | |
CN114757176A (en) | Method for obtaining target intention recognition model and intention recognition method | |
JP6994289B2 (en) | Programs, devices and methods for creating dialogue scenarios according to character attributes | |
CN111639162A (en) | Information interaction method and device, electronic equipment and storage medium | |
CN110457454A (en) | A kind of dialogue method, server, conversational system and storage medium | |
CN108306813B (en) | Session message processing method, server and client | |
CN112507139A (en) | Knowledge graph-based question-answering method, system, equipment and storage medium | |
CN114244795B (en) | Information pushing method, device, equipment and medium | |
WO2021211300A1 (en) | System and method for summerization of customer interaction | |
CN112685551A (en) | Corpus construction method, chat robot conversation method and related device | |
CN117370512A (en) | Method, device, equipment and storage medium for replying to dialogue | |
CN111787042A (en) | Method and device for pushing information | |
CN112559699A (en) | Information interaction method, device and equipment | |
CN109683727A (en) | A kind of data processing method and device | |
CN111241236B (en) | Task-oriented question-answering method, system, electronic device and readable storage medium | |
CN114969250A (en) | Man-machine conversation generation method and device, electronic equipment and storage medium | |
CN114201589A (en) | Dialogue method, dialogue device, dialogue equipment and storage medium |
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 |