CN119128120A - Consultation retrieval method and system based on demand tag configuration - Google Patents
Consultation retrieval method and system based on demand tag configuration Download PDFInfo
- Publication number
- CN119128120A CN119128120A CN202411620755.1A CN202411620755A CN119128120A CN 119128120 A CN119128120 A CN 119128120A CN 202411620755 A CN202411620755 A CN 202411620755A CN 119128120 A CN119128120 A CN 119128120A
- Authority
- CN
- China
- Prior art keywords
- consultation
- retrieval
- priorities
- demand
- library
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a consultation retrieval method and a system based on demand label configuration, and relates to the technical field of data retrieval processing, wherein the method comprises the steps of identifying and classifying a consultation sentence library based on semantic identification; the method comprises the steps of obtaining an emotion tag library and a customer tag library based on emotion recognition tags and customer value recognition tags, analyzing consultation retrieval priorities of demand tags, obtaining historical user consultation record data, matching the consultation priorities, randomly adjusting the consultation priorities, matching the historical consultation data to calculate consultation retrieval fitness, and carrying out consultation retrieval according to an optimal consultation retrieval scheme. The method solves the technical problems of inaccurate search results and poor search information due to the fact that the user can not accurately analyze the requirement of acquiring information in the existing consultation search, and ensures that the search accuracy and efficiency are low, achieves more accurate and efficient information search, and achieves the technical effects of improving the efficiency and accuracy of the consultation search.
Description
Technical Field
The application relates to the technical field of data retrieval processing, in particular to a consultation retrieval method and a consultation retrieval system based on demand label configuration.
Background
In the age of information explosion, the difficulty of people to acquire and screen required information is gradually increased, no matter whether an individual user or an enterprise organization needs a high-efficiency and accurate retrieval tool to help the individual user or the enterprise organization to quickly find information matched with own requirements, and the consultation retrieval is a special information retrieval mode, and is characterized in that the consultation intention of the user is understood, the accurate matching is carried out according to the requirements of the user, the consultation problem of the user can be classified into a corresponding theme or field through the user retrieval requirements, so that the quick positioning and screening of related information are realized, but the traditional consultation retrieval is carried out by either consulting a catalog, indexing and the like or carrying out remote access retrieval through the Internet, so that the retrieval efficiency is low, the matching degree and the accuracy of retrieval results are not high, and the requirements of the user on acquiring information at any time and any place can not be accurately and quickly met due to the fact that the region and time are limited.
Therefore, in the related technology of the consultation and retrieval at the present stage, the technical problems of inaccurate retrieval results and redundant retrieval information, and low retrieval accuracy and efficiency caused by the fact that the requirement of a user for acquiring information cannot be accurately resolved exist.
Disclosure of Invention
According to the consultation retrieval method and system based on the demand label configuration, the technical means of semantic recognition, emotion classification, preset retrieval rules and the like are adopted, so that the technical problems that the retrieval result is inaccurate and the retrieval information is redundant due to the fact that the user's acquired information needs cannot be accurately analyzed in the existing consultation retrieval are solved, the retrieval accuracy and efficiency are low are solved, more accurate and efficient information retrieval is achieved, and the technical effects of improving the efficiency and accuracy of the consultation retrieval are achieved.
The application provides a consultation retrieval method based on demand label configuration, which comprises the steps of identifying and classifying a consultation statement library based on semantic identification to obtain a consultation question library, obtaining a mood label library and a customer label library based on mood identification marks and customer value identification marks, combining the consultation question library to obtain a demand label library, wherein the demand label library comprises a plurality of demand labels, analyzing consultation retrieval priorities of the demand labels to obtain a plurality of consultation priorities as a basic consultation retrieval scheme, obtaining historical user consultation record data, matching according to the plurality of consultation priorities to obtain matching historical consultation data, calculating to obtain first consultation retrieval adaptability, randomly adjusting the plurality of consultation priorities in the basic consultation retrieval scheme, matching the historical consultation data to calculate consultation retrieval adaptability, optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme, and carrying out consultation retrieval on consultation demands provided with the plurality of demand labels according to the optimal consultation retrieval scheme.
In a possible implementation manner, the method comprises the steps of identifying and classifying a consultation statement library based on semantic identification to obtain a consultation statement library, acquiring a historical consultation statement set based on a historical consultation data record to construct the consultation statement library, acquiring a sample consultation statement set, identifying consultation questions in the sample consultation statement to obtain a sample consultation question set, constructing a consultation question identifier based on semantic identification by adopting the sample consultation statement set and the sample consultation question set, identifying the consultation statement library to obtain a plurality of consultation questions, and classifying to obtain the consultation statement library.
In a possible implementation manner, based on the emotion recognition mark and the customer value recognition mark, an emotion tag library and a customer tag library are obtained, and the method is combined with the consultation question library, and comprises the steps of obtaining a sample consultation audio set, identifying user emotion in the sample consultation audio to obtain a sample user emotion set, adopting the sample consultation audio set and the sample user emotion set, based on audio recognition, constructing an emotion recognizer, recognizing the consultation sentence library to obtain a plurality of user emotions, classifying to obtain an emotion tag library, based on user service categories of a plurality of users, constructing a user tag library, and combining the consultation question library, the emotion tag library and the customer tag library to obtain the demand tag library.
In a possible implementation manner, the consultation retrieval priorities of the plurality of demand labels are analyzed to obtain a plurality of consultation priorities, the following processes are further performed, namely the plurality of demand labels in the demand label library are ordered and assigned according to the order of occurrence probability of consultation problems from large to small to obtain a plurality of problem priorities, wherein the more front the ordering is, the greater the problem priorities are, the plurality of demand labels in the demand label library are ordered and assigned according to the order of occurrence probability of emotion labels from small to large to obtain a plurality of emotion priorities, the plurality of demand labels in the demand label library are ordered and assigned according to the order of customer values from large to small to obtain a plurality of user priorities, and the plurality of problem priorities, the plurality of emotion priorities and the plurality of customer priorities are weighted and calculated to obtain the plurality of consultation priorities.
In a possible implementation manner, historical user consultation record data are obtained, matching is carried out according to the plurality of consultation priorities, matching historical consultation data are obtained, first consultation retrieval fitness is calculated and obtained, the historical user consultation record data are obtained, the historical user consultation record data comprise a plurality of consultation retrieval data sets, each consultation retrieval data set comprises a plurality of requirement labels of a plurality of consultations and a plurality of consultation retrieval orders, the plurality of consultation retrieval data sets are matched according to the plurality of consultation priorities, a plurality of first matching consultation retrieval data sets with the consultation retrieval orders conforming to the plurality of consultation priorities are obtained, and the first consultation retrieval fitness of the basic consultation retrieval scheme is calculated and obtained according to a plurality of user evaluation data sets of the plurality of first matching consultation retrieval data sets, wherein each user evaluation data comprises an evaluation grade.
In a possible implementation manner, according to the multiple user evaluation data sets of the multiple first matching consultation retrieval data sets, a first consultation retrieval fitness of the basic consultation retrieval scheme is obtained through calculation, and a consultation retrieval function for calculating the fitness of the consultation retrieval scheme is constructed, wherein the following formula is:
;
Wherein the CSC is the consultation retrieval fitness, M is the number of all consultation requirements in the plurality of first matched consultation retrieval data sets, Weights assigned to the sizes of consultation priorities according to the ith consultation requirement,For the user rating in the user rating data of the i-th consultation requirement,And according to the consultation retrieval function, combining the consultation priorities of the consultation requests in the plurality of first matched consultation retrieval data sets and the plurality of user evaluation data sets to calculate and obtain the first consultation retrieval adaptability.
In a possible implementation manner, the method comprises the steps of randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, matching historical consultation data to calculate consultation retrieval adaptability, optimizing the basic consultation retrieval scheme, randomly selecting P consultation priorities of P demand labels in the demand label library, randomly adjusting to obtain P adjustment consultation priorities, obtaining an adjustment basic consultation retrieval scheme, wherein P is an integer greater than 1 and less than the number of all demand labels, calculating consultation retrieval adaptability based on the adjustment basic consultation retrieval scheme and matching historical consultation data to obtain second consultation retrieval adaptability, judging whether the second consultation retrieval adaptability is greater than the first consultation retrieval adaptability, if so, continuing to randomly adjust the P consultation priorities of the P demand labels, if not, using the P demand labels as tabu labels, reselecting the priority of other P demand labels, randomly adjusting and optimizing the consultation priorities, and stopping the tabu labels after the tabu times, and outputting the optimized consultation retrieval adaptability until the optimized consultation retrieval adaptability is the optimal consultation retrieval adaptability.
The application also provides a consultation retrieval system based on the demand label configuration, which comprises:
The semantic recognition and classification module is used for recognizing and classifying the consultation sentence library based on semantic recognition to obtain a consultation inquiry question library;
The demand label library obtaining module is used for obtaining an emotion label library and a customer label library based on emotion recognition marks and customer value recognition marks, combining the consultation problem library to obtain a demand label library, wherein the demand label library comprises a plurality of demand labels;
The retrieval priority analysis module is used for analyzing the consultation retrieval priorities of the plurality of demand labels to obtain a plurality of consultation priorities which are used as a basic consultation retrieval scheme;
The first consultation retrieval adaptability calculation module is used for acquiring historical user consultation record data, matching according to the plurality of consultation priorities, acquiring matched historical consultation data and calculating and acquiring first consultation retrieval adaptability;
the consultation retrieval scheme optimizing module is used for randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, calculating consultation retrieval fitness by matching historical consultation data, and optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme;
And the consultation retrieval module is used for performing consultation retrieval on the consultation demands configured with the plurality of demand labels according to the optimal consultation retrieval scheme.
The consultation retrieval method and the system based on the demand label configuration are used for identifying and classifying a consultation statement library based on semantic identification, obtaining an emotion label library and a customer label library based on emotion identification marks and customer value identification marks, analyzing consultation retrieval priority of demand labels, obtaining historical user consultation record data, matching according to a plurality of consultation priorities, randomly adjusting the plurality of consultation priorities, matching the historical consultation data to calculate consultation retrieval fitness, and carrying out consultation retrieval according to an optimal consultation retrieval scheme. The method solves the technical problems of inaccurate search results and poor search information due to the fact that the user can not accurately analyze the requirement of acquiring information in the existing consultation search, and ensures that the search accuracy and efficiency are low, achieves more accurate and efficient information search, and achieves the technical effects of improving the efficiency and accuracy of the consultation search.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following will briefly describe the drawings of the embodiments of the present disclosure, in which flowcharts are used to illustrate operations performed by a system according to embodiments of the present disclosure. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
FIG. 1 is a schematic flow chart of a method for consulting and retrieving based on a demand label configuration according to an embodiment of the present application;
Fig. 2 is a schematic diagram of a query retrieval system configured based on a requirement label according to an embodiment of the present application.
Reference numerals illustrate the semantic recognition and classification module 10, the demand label library obtaining module 20, the retrieval priority analysis module 30, the first consultation retrieval adaptability calculation module 40, the consultation retrieval scheme optimization module 50 and the consultation retrieval module 60.
Detailed Description
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail with reference to the accompanying drawings, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, the term "first\second" being referred to merely as distinguishing between similar objects and not representing a particular ordering for the objects. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules that may not be expressly listed or inherent to such process, method, article, or apparatus, and unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used herein is for the purpose of describing embodiments of the application only.
The embodiment of the application provides a consultation retrieval method based on demand label configuration, as shown in fig. 1, the method comprises the following steps:
The consultation retrieval method based on the demand label configuration combines the advantages of the modern information technology, can quickly and accurately understand the consultation intention of the user, configure the demand label, accurately match the configuration label, classify the consultation problem of the user into the field and the theme corresponding to the demand label, realize the quick positioning and screening of related information, provide more personalized information recommendation and solution for the user, meet the increasing information retrieval requirement of the user, help the user solve various problems such as academic research, business decision, life service and the like, and realize more accurate and efficient information consultation retrieval.
Step S100, based on semantic recognition, the consultation sentence library is recognized and classified, and the consultation inquiry question library is obtained. The method comprises the steps of carrying out deep analysis and understanding on a large number of consultation sentences by utilizing a semantic recognition technology, extracting semantic meanings of the sentences, classifying and arranging the sentences according to the meanings, and finally forming a structured consultation question bank, specifically, carrying out deep analysis on grammar structures, vocabulary meanings, context relations and the like of the consultation sentences by semantic recognition, understanding specific intentions and meanings expressed by the sentences, classifying similar consultation sentences into the same theme or category according to the meanings, forming different consultation question sets, and forming the consultation question bank.
In a possible implementation manner, step S100 further includes step S110, based on the historical consultation data record, obtaining a historical consultation statement set, and constructing a consultation statement library. Extracting consultation sentences from the past consultation data records by collecting and analyzing the past consultation data records, and arranging the sentences into a set to further construct a consultation sentence library for subsequent analysis or processing, specifically, the historical consultation data records comprise information of collecting the past consultation records from various sources (such as a customer service center, an online consultation platform, an e-mail and the like), such as problems of customers, consultation contents, context of conversations and the like; the set of historical advisory statements is typically a set of questions or expressions posed by the customer. Step S120 is also included, a sample consultation statement set is obtained, consultation questions in the sample consultation statement are identified, and the sample consultation question set is obtained. Representative consultation sentences are screened from the existing consultation records or databases, the sample consultation sentences can cover the consultation problems of different fields, different topics and different expression modes to form a set, each sample consultation sentence is required to be deeply analyzed and understood as a sample consultation sentence set so as to accurately identify the consultation problems therein, and after identification, a sample consultation problem set containing a plurality of consultation problems is obtained, and each problem is required to be associated with the original consultation sentence. The method also comprises a step S130 of constructing a consultation problem identifier by adopting the sample consultation statement set and the sample consultation problem set based on semantic recognition, identifying the consultation statement library to obtain a plurality of consultation problems, and classifying to obtain a consultation inquiry question library. According to semantic recognition, a sample consultation sentence set is analyzed, keywords and phrases in sentences and relations among the keywords and phrases are extracted, a consultation problem recognizer can be constructed by utilizing the sample consultation sentence set and a corresponding sample consultation problem set, consultation problems in new consultation sentences can be automatically recognized by learning characteristics and rules in sample data and are applied to a consultation sentence library, consultation problems in each sentence in the consultation sentence library can be automatically extracted by a model through recognition of each sentence in the consultation sentence library, the consultation problems can be associated with the corresponding sentences, a plurality of consultation problems can be quickly and accurately obtained from a large number of consultation sentences, and the recognized consultation problems are classified and sorted to form the consultation inquiry question library.
Step S200, based on the emotion recognition marks and the customer value recognition marks, an emotion tag library and a customer tag library are obtained, and a demand tag library is obtained by combining the consultation problem library, wherein the demand tag library comprises a plurality of demand tags. The emotion recognition mark is used for capturing and analyzing emotion tendencies in consultation sentences or questions, such as positive, negative, neutral and the like, emotion recognition of the consultation sentences can form an emotion tag library, the emotion tag library comprises various emotion types and corresponding marks thereof, the customer value recognition mark is used for evaluating the value of a customer based on factors such as purchase records, liveness, feedback quality and the like of the customer, specifically, the user value of each consultation sentence is recognized according to the semantics, corresponding tags are marked according to the payment amount or member grade of the user, a plurality of emotion tags and customer tags are obtained, an emotion tag library and a customer tag library are respectively formed, the consultation questions, the emotion tags and the customer tags are combined to form a demand tag library, and each demand tag comprises contents of the consultation questions, emotion states of the customer and value information of the customer and can reflect demands of the customer more comprehensively. Through the demand label library, enterprises can more accurately understand the demands and the expectations of clients, and personalized services or solutions are provided for different clients.
In a possible implementation manner, step S200 further includes step S210, obtaining a sample consultation audio set, and identifying a user emotion in the sample consultation audio, and obtaining a sample user emotion set. The sample consultation audio frequency set refers to an audio frequency data set containing user consultation contents collected from various channels (such as telephone recording, voice chat tools, customer service hotlines and the like), different consultation scenes, user groups and emotion expressions are carried out, each sample consultation audio frequency is used for analyzing the emotion of the user and judging the emotion type expressed by the user, such as happiness, gas generation, sadness, surprise, calm and the like, the labeling of corresponding emotion labels is carried out, and all the sample consultation audio frequencies labeled with the emotion labels and the emotion labels corresponding to the sample consultation audio frequencies are arranged into one set, namely a sample user emotion set. And step S220, adopting the sample consultation audio set and the sample user emotion set, constructing an emotion recognizer based on audio recognition, recognizing the consultation statement library, obtaining a plurality of user emotions, and classifying to obtain an emotion tag library. The collected and marked sample consultation audio set and sample user emotion set are utilized to train an emotion recognizer, the emotion of the user can be automatically recognized according to the characteristics in the audio signals, the emotion recognizer is applied to a consultation sentence library after being constructed, specifically, data in the consultation sentence library is usually in a text form, the text is required to be converted into voice, the emotion recognizer is used for recognizing the audio, so that a plurality of user emotions are obtained, and finally, the recognized user emotions are classified and arranged to form an emotion label library. Step S230 is also included, based on the user service categories of the various users, a user tag library is constructed. Classifying, integrating and tagging user data according to user service categories of users, and constructing a user tag library, wherein the user tag library comprises basic information (such as age, gender, occupation and the like) of the users, behavior data (such as access frequency, use duration, purchase records and the like) and feedback data (such as evaluation, complaints, suggestions and the like). And step S240, combining the consultation problem library, the emotion tag library and the client tag library to obtain the demand tag library. Integrating the obtained consultation inquiry question library, the emotion tag library and the customer tag library to finally obtain the demand tag library.
And step S300, analyzing the consultation retrieval priorities of the plurality of demand labels to obtain a plurality of consultation priorities as a basic consultation retrieval scheme. The method comprises the steps of analyzing consultation retrieval priority of a plurality of demand labels, generally, assigning a priority score to each demand label based on evaluation standards such as business value, emergency degree, implementation difficulty, user expectation and the like of the demand, referring to historical data, user feedback, business targets and the like, sorting the plurality of demand labels based on the priority scores to form a plurality of consultation priorities, specifically, after a user presents a consultation problem, retrieving corresponding basic answers through artificial intelligence to answer, and the consultation priority is similar to that of a customer service robot, when a plurality of users initiate consultation simultaneously, the consultation priority is the sequential priority of retrieving the consultation problem, the larger the priority is, the consultation retrieval is firstly performed, the waiting time is short, otherwise, the consultation retrieval is performed later, the waiting time is long, and the consultation priority is for example, 10 grades of 1-10.
In a possible implementation manner, step S300 further includes step S310, sorting and assigning priorities to the plurality of demand labels in the demand label library according to the order of occurrence probability of the consultation questions from high to low, so as to obtain a plurality of question priorities, where the more front the sorting is, the greater the question priorities are. And step S320, sorting and distributing the priorities of the plurality of demand labels in the demand label library according to the sequence of the occurrence probability of the emotion labels from small to large, so as to obtain a plurality of emotion priorities. The demand labels are ordered in order of the probability of occurrence of the emotion labels from small to large, and the demands which are easier to cause negative emotion are identified so as to give higher priority to subsequent services, and the negative emotion often means that the satisfaction degree of users is lower, and the problems or demands of the users need to be solved more quickly and accurately, so that each demand label is assigned with an emotion priority, for example, the emotion with lower probability of occurrence (such as positive, satisfaction and the like) can be assigned with lower priority, and the emotion with higher probability of occurrence (such as negative, dissatisfaction and the like) can be assigned with higher priority. And step S330, sorting and distributing the priorities of the plurality of demand labels in the demand label library according to the order of the customer values from high to low, so as to obtain a plurality of user priorities. And step S340, weighting calculation is carried out on the plurality of question priorities, the plurality of emotion priorities and the plurality of client priorities to obtain the plurality of consultation priorities. The weight of each priority reflects the importance degree of different priorities in the consultation retrieval process, for example, the problem priority may reflect the urgency and importance of the consultation problem, the emotion priority may reflect the emotion state and satisfaction of the client, the client priority may be based on the value or importance of the client, for each consultation case, the corresponding problem priority, emotion priority and client priority are multiplied by the weight respectively, the priority after comprehensively considering a plurality of factors is reflected, finally, the consultation cases are ordered according to the weighted values, so that a plurality of consultation priorities are obtained and used as the basis of the consultation retrieval process, and the more important, urgent or valuable cases can be preferentially processed when a large number of consultations are processed.
Step S400, historical user consultation record data are obtained, matching is conducted according to the consultation priorities, matching historical consultation data are obtained, and first consultation retrieval fitness is calculated and obtained. The method comprises the steps of obtaining historical user consultation record data, matching the historical consultation record data according to a plurality of determined consultation priorities (obtained by weighting calculation based on problem priorities, emotion priorities and client priorities), obtaining a group of historical consultation data matched with current consultation, calculating first consultation retrieval fitness, specifically, matching the historical user consultation record data according to a plurality of time periods in historical time, according to the consultation retrieval sequence in the historical time, outputting consultation data with the same sequence and priority, for example, the priority of the first consultation request for consultation retrieval is the maximum value of the priorities of all consultation requests in the time period, and so on, and calculating the fitness of a basic consultation retrieval scheme.
In a possible implementation manner, step S400 further includes step S410 of obtaining historical user consultation record data, where the historical user consultation record data includes a plurality of consultation retrieval data sets, and each consultation retrieval data set includes a plurality of requirement labels of a plurality of consultations and a plurality of consultation retrieval orders. The historical user consultation record data refers to user consultation data accumulated in a past period and comprises a large amount of user consultation information, such as consultation time, consultation contents, consultation results and the like, the consultation retrieval data sets are subsets which are divided from the historical user consultation record data, each consultation retrieval data set is focused on a specific consultation theme or problem type, each consultation retrieval data set comprises a plurality of requirement labels of multiple consultations and a plurality of consultation retrieval orders, the consultation retrieval orders refer to the sequence of the user for retrieval or inquiry in the multiple consultations, and the sequence is influenced by a plurality of factors, such as the urgency of user requirements, the complexity of the consultation contents, the satisfaction degree of the historical consultation results and the like. And step S420, matching the plurality of consultation retrieval data sets according to the plurality of consultation priorities to obtain a plurality of first matched consultation retrieval data sets with the consultation retrieval orders conforming to the plurality of consultation priorities. Further comprising step S430, calculating a first query search fitness of the basic query search scheme according to a plurality of user evaluation data sets of the plurality of first matching query search data sets, wherein each user evaluation data includes an evaluation level. The effectiveness of the basic consultation retrieval scheme is evaluated by combining evaluation data of the matched consultation retrieval data sets of users, wherein each user evaluation data comprises an evaluation grade reflecting satisfaction and acceptance of the users to the retrieval result, the user evaluation data is the grade of the users after consultation, for example, 10 grades of 1-10, the user evaluation data of the basic consultation retrieval scheme is calculated according to direct feedback of users, such as scoring, grading, commenting and the like, of the first matched consultation retrieval data sets, and the first consultation retrieval fitness of the basic consultation retrieval scheme is calculated according to the average evaluation scores.
In a possible implementation manner, step S430 further includes step S431, constructing a consulting search function for calculating a suitability of the consulting search scheme, where:
;
Wherein the CSC is the consultation retrieval fitness, M is the number of all consultation requirements in the plurality of first matched consultation retrieval data sets, Weights assigned to the sizes of consultation priorities according to the ith consultation requirement,For the user rating in the user rating data of the i-th consultation requirement,Consultation priority for the ith consultation requirement. Wherein, the larger the consultation priority is, the larger the corresponding assigned weight is,The greater the value of (2). And step S432, according to the consultation retrieval function, combining the consultation priorities of the consultation requests in the first matched consultation retrieval data sets and the user evaluation data sets, and calculating to obtain first consultation retrieval fitness. The consultation retrieval function is generally used for quickly and accurately finding historical consultation data which is most matched with the current consultation request in a large amount of consultation data, combining consultation priorities of a plurality of consultation requests in a plurality of first matched consultation retrieval data sets and a plurality of user evaluation data sets according to the consultation retrieval function, specifically, weighting the user evaluation data according to the consultation priorities so that the consultation request with higher priority occupies larger weight in fitness calculation, and calculating first consultation retrieval fitness according to the matching degree returned by the consultation retrieval function and the user evaluation data to reflect comprehensive performance of the current consultation retrieval scheme in terms of matching degree and user satisfaction.
And S500, randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, matching historical consultation data to calculate consultation retrieval fitness, and optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme. The method comprises the steps of randomly adjusting consultation priorities in a basic consultation retrieval scheme, generating diversified candidate schemes in a search space, using the randomly adjusted consultation priorities to match historical consultation data, calculating corresponding consultation retrieval fitness, evaluating the advantages and disadvantages of the current retrieval scheme according to the calculated consultation retrieval fitness, if the fitness is improved, indicating that the random adjustment has a positive effect, keeping the adjustment, otherwise, if the fitness is reduced, returning to the previous scheme, repeating, and gradually optimizing the basic consultation retrieval scheme by continuously randomly adjusting, matching historical data, calculating the fitness and evaluating the effect, wherein the retrieval scheme gradually converges to a better state along with the increase of iteration times, namely the optimal consultation retrieval scheme.
In a possible implementation manner, step S500 further includes step S510, where P consultation priorities of P demand labels are randomly selected in the demand label library to perform random adjustment, P adjustment consultation priorities are obtained, an adjustment basic consultation retrieval scheme is obtained, and P is an integer greater than 1 and less than the total number of demand labels. P demand labels are randomly selected from a demand label library, the corresponding consultation priorities of the P demand labels are randomly adjusted according to the P demand labels, namely, the priority order or the numerical value of the demand labels is changed, P adjustment consultation priorities are obtained, and an adjustment basic consultation retrieval scheme is formed based on the adjusted priorities, so that the overall retrieval effect is optimized by changing the priorities of part of demand labels. And step S520, calculating the consultation retrieval fitness based on the adjustment basic consultation retrieval scheme and matching the historical consultation data to obtain a second consultation retrieval fitness. The second consultation retrieval adaptability is the consultation retrieval adaptability corresponding to the adjustment basic consultation retrieval scheme. And step S530, judging whether the second consultation retrieval adaptability is larger than the first consultation retrieval adaptability, if so, continuing to randomly adjust the P consultation priorities of the P demand labels, if not, taking the P demand labels as tabu labels, reselecting the consultation priorities of other P demand labels for random adjustment optimization, and after the tabu times, performing forbidden label resolution. If the second consultation retrieval adaptability is smaller than the first consultation retrieval adaptability, the P demand labels are used as tabu labels, the consultation priorities of other demand labels are reselected for random adjustment optimization, the tabu labels are labels which do not adjust the priorities within the tabu times, so that a better solution can be found more quickly, and after the tabu times are reached, the tabu labels are forbidden, and the consultation priorities can be adjusted. And step S530, continuing to optimize until convergence, and outputting the consultation retrieval scheme with the largest consultation retrieval adaptability as the optimal consultation retrieval scheme. Specifically, in the current basic consultation retrieval scheme, the consultation priorities of the P demand labels are adjusted to generate new candidate consultation retrieval schemes, the new candidate consultation retrieval schemes are used for matching historical consultation data, the consultation retrieval fitness of the new schemes is calculated based on user evaluation data or other indexes, the consultation retrieval fitness of the new schemes is compared with the fitness of the current optimal scheme, if the fitness of the new schemes is higher, the new schemes are replaced with the current optimal scheme, optimization is repeatedly performed, new candidate schemes are continuously generated and evaluated until certain convergence conditions are met, for example, the improvement of the fitness reaches a tiny threshold value or the iterative times reach preset times, and the consultation retrieval scheme with the largest consultation retrieval fitness obtained in the iterative process is output as the optimal consultation retrieval scheme.
Step S600, according to the optimal consultation retrieval scheme, consultation retrieval is carried out on the consultation demands configured with the plurality of demand labels. When a user puts forward a consultation demand, the system executes consultation retrieval operation according to the determined optimal consultation retrieval scheme to find the most relevant and valuable consultation data, specifically, analyzes the consultation demand of the user, extracts key information in the consultation demand, converts the key information into demand labels, determines the consultation priority of each demand label according to the optimal consultation retrieval scheme obtained through the optimization process, retrieves in historical consultation data according to the determined demand labels and the corresponding consultation priority, sorts the retrieved historical consultation data based on the matching degree, and preferentially displays the result with the highest matching degree and the highest value with the current consultation demand.
In the above, the consultation searching method based on the demand label configuration according to the embodiment of the present invention is described in detail with reference to fig. 1. Next, a counseling retrieval system based on a demand label configuration according to an embodiment of the present invention will be described with reference to fig. 2.
According to the consultation retrieval system based on the requirement label configuration, which is disclosed by the embodiment of the invention, the technical problems that the retrieval result is inaccurate and the retrieval information is redundant due to the fact that the requirement of a user for obtaining information cannot be accurately analyzed in the existing consultation retrieval are solved, so that the retrieval accuracy and the retrieval efficiency are low are solved, the more accurate and efficient information retrieval is realized, and the technical effects of improving the efficiency and the accuracy of the consultation retrieval are achieved. The consultation retrieval system based on the requirement label configuration comprises a semantic recognition classification module 10, a requirement label library obtaining module 20, a retrieval priority analysis module 30, a first consultation retrieval adaptability calculation module 40, a consultation retrieval scheme optimization module 50 and a consultation retrieval module 60.
The semantic recognition and classification module 10 is used for recognizing and classifying the consultation sentence library based on semantic recognition to obtain a consultation inquiry question library;
The demand label library obtaining module 20 is configured to obtain an emotion label library and a customer label library based on emotion recognition marks and customer value recognition marks, and combine the consultation problem library to obtain a demand label library, where the demand label library includes a plurality of demand labels;
The search priority analysis module 30 is configured to analyze the consultation search priorities of the plurality of requirement labels, and obtain a plurality of consultation priorities as a basic consultation search scheme;
the first consultation retrieval adaptability calculation module 40 is used for acquiring historical user consultation record data, matching according to the plurality of consultation priorities, acquiring matched historical consultation data and calculating and acquiring first consultation retrieval adaptability;
The consultation retrieval scheme optimizing module 50 is used for randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, matching historical consultation data to calculate consultation retrieval fitness, and optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme;
The consultation retrieval module 60 is configured to perform consultation retrieval on the consultation demands configured with the plurality of demand labels according to the optimal consultation retrieval scheme by the consultation retrieval module 60.
Next, the specific configuration of the semantic recognition classification module 10 will be described in detail. The semantic recognition classification module 10 further comprises the steps of obtaining a historical consultation statement set based on a historical consultation data record, constructing a consultation statement library, obtaining a sample consultation statement set, identifying consultation questions in the sample consultation statement set, obtaining a sample consultation question set, constructing a consultation question recognizer based on semantic recognition by adopting the sample consultation statement set and the sample consultation question set, recognizing the consultation statement library, obtaining a plurality of consultation questions, and classifying to obtain the consultation question library.
Next, the specific configuration of the demand tag library obtaining module 20 will be described in detail. The demand tag library obtaining module 20 may further obtain a sample consultation audio set, identify a user emotion in the sample consultation audio set, obtain a sample user emotion set, use the sample consultation audio set and the sample user emotion set, construct an emotion recognizer based on audio recognition, recognize the consultation statement library, obtain a plurality of user emotions, classify to obtain an emotion tag library, construct a user tag library based on user service categories of a plurality of users, and combine the consultation question library, the emotion tag library and the client tag library to obtain the demand tag library.
Next, the specific configuration of the retrieval priority analysis module 30 will be described in detail. The search priority analysis module 30 may further include sorting and priority allocation of the plurality of demand labels in the demand label library according to the order of the occurrence probability of the consultation problem from large to small to obtain a plurality of problem priorities, wherein the more the problem priorities are sorted, sorting and priority allocation of the plurality of demand labels in the demand label library according to the order of the occurrence probability of the emotion labels from small to large to obtain a plurality of emotion priorities, sorting and priority allocation of the plurality of demand labels in the demand label library according to the order of the client value from large to small to obtain a plurality of user priorities, and weighting calculation of the plurality of problem priorities, the plurality of emotion priorities and the plurality of client priorities to obtain the plurality of consultation priorities.
Next, the specific configuration of the first consulting retrieval adaptability calculation module 40 will be described in detail. The first query search fitness calculation module 40 still further includes obtaining historical user query record data, wherein the historical user query record data includes a plurality of query search data sets, each query search data set includes a plurality of requirement labels for multiple queries and a plurality of query search orders, matching the plurality of query search data sets according to the plurality of query priorities to obtain a plurality of first matched query search data sets with the plurality of query priorities, and calculating a first query search fitness for obtaining the basic query search scheme according to a plurality of user evaluation data sets of the plurality of first matched query search data sets, wherein each user evaluation data includes an evaluation level.
Next, the specific configuration of the first consulting retrieval adaptability calculation module 40 will be described in further detail. The first counsel retrieval fitness calculating module 40 further includes constructing a counsel retrieval function that calculates a fitness of a counsel retrieval scheme, as follows:
;
Wherein the CSC is the consultation retrieval fitness, M is the number of all consultation requirements in the plurality of first matched consultation retrieval data sets, Weights assigned to the sizes of consultation priorities according to the ith consultation requirement,For the user rating in the user rating data of the i-th consultation requirement,And according to the consultation retrieval function, combining the consultation priorities of the consultation requests in the plurality of first matched consultation retrieval data sets and the plurality of user evaluation data sets to calculate and obtain the first consultation retrieval adaptability.
Next, the specific configuration of the consulting retrieval scheme optimization module 50 will be described in detail. The consultation retrieval scheme optimizing module 50 may further include randomly selecting P consultation priorities of P demand labels in the demand label library to perform random adjustment to obtain P adjustment consultation priorities, obtaining adjustment basic consultation retrieval schemes, wherein P is an integer greater than 1 and less than the number of all demand labels, calculating consultation retrieval fitness based on the adjustment basic consultation retrieval schemes and matching historical consultation data to obtain second consultation retrieval fitness, judging whether the second consultation retrieval fitness is greater than the first consultation retrieval fitness, if so, continuing to perform random adjustment on the P consultation priorities of the P demand labels, if not, using the P demand labels as tabu labels, reselecting consultation priorities of other P demand labels to perform random adjustment optimization, performing solution prohibition on the tabu labels after the tabu numbers, continuing to perform optimization until convergence, and outputting the consultation retrieval scheme with the maximum consultation retrieval fitness as the optimal consultation retrieval scheme.
The consultation retrieval system based on the demand label configuration provided by the embodiment of the invention can execute the consultation retrieval method based on the demand label configuration provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Although the present application makes various references to certain modules in a system according to an embodiment of the present application, any number of different modules may be used and run on a user terminal and/or a server, and each unit and module included are merely divided according to functional logic, but are not limited to the above-described division, so long as the corresponding functions can be implemented, and in addition, specific names of each functional unit are only for convenience of distinguishing from each other, and are not intended to limit the scope of protection of the present application.
The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application.
Claims (8)
1. The consultation retrieval method based on the demand label configuration is characterized by comprising the following steps:
Based on semantic recognition, recognizing and classifying the consultation sentence library to obtain a consultation inquiry question library;
Based on the emotion recognition marks and the customer value recognition marks, obtaining an emotion tag library and a customer tag library, combining the consultation problem library, and combining to obtain a demand tag library, wherein the demand tag library comprises a plurality of demand tags;
analyzing the consultation retrieval priorities of the plurality of demand labels to obtain a plurality of consultation priorities as a basic consultation retrieval scheme;
acquiring historical user consultation record data, matching according to the plurality of consultation priorities, acquiring matched historical consultation data, and calculating and acquiring first consultation retrieval fitness;
Randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, matching historical consultation data to calculate consultation retrieval fitness, and optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme;
And according to the optimal consultation retrieval scheme, performing consultation retrieval on the consultation demands configured with the plurality of demand labels.
2. The consulting retrieval method based on demand label configuration of claim 1, wherein the consulting statement library is identified and classified based on semantic recognition to obtain a consulting inquiry question library, comprising:
based on the historical consultation data record, acquiring a historical consultation statement set and constructing a consultation statement library;
Based on semantic recognition, adopting the sample consultation statement set and the sample consultation problem set to construct a consultation problem recognizer, recognizing the consultation statement library to obtain a plurality of consultation problems, and classifying to obtain the consultation inquiry question library.
3. The consultation retrieval method based on the demand label configuration of claim 2, wherein based on the emotion recognition mark and the customer value recognition mark, an emotion label library and a customer label library are obtained, and in combination with the consultation question library, comprising:
Acquiring a sample consultation audio set, and identifying the emotion of the user in the sample consultation audio set to acquire a sample user emotion set;
Adopting the sample consultation audio set and the sample user emotion set, constructing an emotion recognizer based on audio recognition, recognizing the consultation statement library to obtain a plurality of user emotions, and classifying to obtain an emotion tag library;
constructing a user tag library based on user service categories of various users;
and combining the consultation problem library, the emotion tag library and the client tag library to obtain the demand tag library.
4. The consultation retrieval method based on the demand label configuration of claim 1, wherein analyzing the consultation retrieval priorities of the plurality of demand labels to obtain a plurality of consultation priorities includes:
sequencing and distributing priorities of a plurality of demand labels in the demand label library according to the sequence of the occurrence probability of the consultation questions from large to small to obtain a plurality of question priorities, wherein the higher the sequencing is, the higher the question priorities are;
Sequencing and distributing priorities of a plurality of demand labels in the demand label library according to the sequence of the occurrence probability of the emotion labels from small to large to obtain a plurality of emotion priorities;
Sequencing and distributing priorities of a plurality of demand labels in the demand label library according to the order of the customer values from large to small to obtain a plurality of user priorities;
And carrying out weighted calculation on the plurality of question priorities, the plurality of emotion priorities and the plurality of client priorities to obtain the plurality of consultation priorities.
5. The consultation retrieval method based on the demand label configuration of claim 1, wherein obtaining historical user consultation record data, matching according to the plurality of consultation priorities, obtaining matching historical consultation data, and calculating to obtain a first consultation retrieval fitness includes:
Acquiring historical user consultation record data, wherein the historical user consultation record data comprises a plurality of consultation retrieval data sets, and each consultation retrieval data set comprises a plurality of requirement labels of a plurality of consultations and a plurality of consultation retrieval orders;
Matching the plurality of consultation retrieval data sets according to the plurality of consultation priorities to obtain a plurality of first matched consultation retrieval data sets with the consultation retrieval orders conforming to the plurality of consultation priorities;
And calculating and obtaining first consultation retrieval fitness of the basic consultation retrieval scheme according to a plurality of user evaluation data sets of the plurality of first matched consultation retrieval data sets, wherein each user evaluation data comprises an evaluation grade.
6. The demand label configuration-based consultation retrieval method of claim 5 characterised in that calculating a first consultation retrieval fitness to obtain said basic consultation retrieval plan from a plurality of user rating data sets of said plurality of first matching consultation retrieval data sets includes:
Constructing a consultation retrieval function for calculating the adaptability of the consultation retrieval scheme, wherein the consultation retrieval function comprises the following formula:
;
Wherein the CSC is the consultation retrieval fitness, M is the number of all consultation requirements in the plurality of first matched consultation retrieval data sets, Weights assigned to the sizes of consultation priorities according to the ith consultation requirement,For the user rating in the user rating data of the i-th consultation requirement,Consultation priority for the ith consultation requirement;
And according to the consultation retrieval function, combining the consultation priorities of the consultation requests in the first matched consultation retrieval data sets and the user evaluation data sets, and calculating to obtain first consultation retrieval fitness.
7. The consultation retrieval method based on the demand label configuration of claim 1, wherein the step of randomly adjusting the plurality of consultation priorities within the basic consultation retrieval scheme and calculating the consultation retrieval fitness matching the historical consultation data, and the step of optimizing the basic consultation retrieval scheme includes:
Randomly selecting P consultation priorities of P demand labels in the demand label library to carry out random adjustment to obtain P adjustment consultation priorities, and obtaining an adjustment basic consultation retrieval scheme, wherein P is an integer greater than 1 and less than the number of all the demand labels;
calculating the consultation retrieval fitness based on the adjustment basic consultation retrieval scheme and matching historical consultation data to obtain a second consultation retrieval fitness;
judging whether the second consultation retrieval fitness is larger than the first consultation retrieval fitness, if so, continuing to randomly adjust the P consultation priorities of the P demand labels, if not, taking the P demand labels as tabu labels, reselecting the consultation priorities of other P demand labels to randomly adjust and optimize, and after the tabu times, performing forbidden elimination on the tabu labels;
And continuing to optimize until convergence, and outputting the consultation retrieval scheme with the maximum consultation retrieval adaptability as the optimal consultation retrieval scheme.
8. A demand label configuration-based consultation retrieval system for implementing the demand label configuration-based consultation retrieval method of any one of claims 1 to 7, the system including:
The semantic recognition and classification module is used for recognizing and classifying the consultation sentence library based on semantic recognition to obtain a consultation inquiry question library;
The demand label library obtaining module is used for obtaining an emotion label library and a customer label library based on emotion recognition marks and customer value recognition marks, combining the consultation problem library to obtain a demand label library, wherein the demand label library comprises a plurality of demand labels;
The retrieval priority analysis module is used for analyzing the consultation retrieval priorities of the plurality of demand labels to obtain a plurality of consultation priorities as a basic consultation retrieval scheme;
The first consultation retrieval adaptability calculation module is used for acquiring historical user consultation record data, matching according to the plurality of consultation priorities, acquiring matched historical consultation data and calculating and acquiring first consultation retrieval adaptability;
the consultation retrieval scheme optimizing module is used for randomly adjusting a plurality of consultation priorities in the basic consultation retrieval scheme, calculating consultation retrieval fitness by matching historical consultation data, and optimizing the basic consultation retrieval scheme until convergence to obtain an optimal consultation retrieval scheme;
And the consultation retrieval module is used for performing consultation retrieval on the consultation demands configured with the plurality of demand labels according to the optimal consultation retrieval scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411620755.1A CN119128120B (en) | 2024-11-14 | 2024-11-14 | Consultation retrieval method and system based on demand label configuration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411620755.1A CN119128120B (en) | 2024-11-14 | 2024-11-14 | Consultation retrieval method and system based on demand label configuration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119128120A true CN119128120A (en) | 2024-12-13 |
| CN119128120B CN119128120B (en) | 2025-04-25 |
Family
ID=93766811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411620755.1A Active CN119128120B (en) | 2024-11-14 | 2024-11-14 | Consultation retrieval method and system based on demand label configuration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119128120B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120523920A (en) * | 2025-07-25 | 2025-08-22 | 广州奥测医学科技有限公司 | Medical knowledge intelligent consulting system based on knowledge graph |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160086114A1 (en) * | 2014-09-22 | 2016-03-24 | Infosys Limited | Service-based consulting framework |
| CN107562866A (en) * | 2017-08-30 | 2018-01-09 | 肖翱 | Matching process and device based on user's request |
| CN116342198A (en) * | 2023-01-06 | 2023-06-27 | 广东丰帆工程咨询有限公司 | Intelligent management method for project consultation data and storage medium |
| CN116521879A (en) * | 2023-05-11 | 2023-08-01 | 中国工商银行股份有限公司 | Customer consultation data processing method and device |
| CN118260398A (en) * | 2024-03-29 | 2024-06-28 | 国家电网有限公司客户服务中心 | A user consultation intelligent answering method, system, device and storage medium |
-
2024
- 2024-11-14 CN CN202411620755.1A patent/CN119128120B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160086114A1 (en) * | 2014-09-22 | 2016-03-24 | Infosys Limited | Service-based consulting framework |
| CN107562866A (en) * | 2017-08-30 | 2018-01-09 | 肖翱 | Matching process and device based on user's request |
| CN116342198A (en) * | 2023-01-06 | 2023-06-27 | 广东丰帆工程咨询有限公司 | Intelligent management method for project consultation data and storage medium |
| CN116521879A (en) * | 2023-05-11 | 2023-08-01 | 中国工商银行股份有限公司 | Customer consultation data processing method and device |
| CN118260398A (en) * | 2024-03-29 | 2024-06-28 | 国家电网有限公司客户服务中心 | A user consultation intelligent answering method, system, device and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120523920A (en) * | 2025-07-25 | 2025-08-22 | 广州奥测医学科技有限公司 | Medical knowledge intelligent consulting system based on knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119128120B (en) | 2025-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN119474280B (en) | Intelligent customer service system based on AI large model | |
| CN108052583B (en) | E-commerce ontology construction method | |
| US6185531B1 (en) | Topic indexing method | |
| CN117668205B (en) | Smart logistics customer service processing method, system, equipment and storage medium | |
| CN107729468B (en) | Answer extraction method and system based on deep learning | |
| US8135579B2 (en) | Method of analyzing conversational transcripts | |
| US20040249808A1 (en) | Query expansion using query logs | |
| US20110093268A1 (en) | Apparatus and method for analysis of language model changes | |
| CN113505606B (en) | Training information acquisition method and device, electronic equipment and storage medium | |
| CN119719312B (en) | Intelligent government affair question-answering method, device, equipment and storage medium | |
| US8781880B2 (en) | System, method and apparatus for voice analytics of recorded audio | |
| CN110516057B (en) | Petition question answering method and device | |
| CN117972160B (en) | Multi-mode information processing method and device | |
| CN119128120B (en) | Consultation retrieval method and system based on demand label configuration | |
| CN115577080A (en) | A question reply matching method, system, server and storage medium | |
| CN111611358A (en) | Information interaction method and device, electronic equipment and storage medium | |
| CN119295088A (en) | Customer service method, service device and computer equipment | |
| US20230108518A1 (en) | Information processing apparatus, information processing method, and program | |
| KR20200143989A (en) | The person in charge assignment system and method based on type classification for customer inquiries | |
| CN116975260A (en) | Complaint work order processing methods, devices, equipment and media based on semantic mining | |
| CN118861438A (en) | A digital consulting management method and system for engineering information based on the Internet of Things | |
| CN116860950A (en) | Method and system for updating corpus of term conversation robot | |
| CN113051934A (en) | User question and sentence mixing system for processing semantic similarity and word order similarity | |
| CN119226473B (en) | Customer service answering intelligent prompt method and system based on speech data | |
| CN116109106B (en) | Intelligent processing method, device and storage medium for work order data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20250604 Address after: 710000 Shaanxi Province, Xi'an City, Yanta District, Qingsong Road No. 58 (Xiaoxi'eryu Dao), Building 1, Room 10701 Patentee after: Xi'an Hefang Information Technology Co.,Ltd. Country or region after: China Address before: 226000 Jiangsu Province Nantong City High-tech Industrial Development Zone New Century Avenue 266 Jianghai Zhihui Park A1 Building 6th Floor 624 Room Patentee before: Nantong Huashidai Information Technology Co.,Ltd. Country or region before: China |