CN114357126A - Intelligent question-answering system - Google Patents
Intelligent question-answering system Download PDFInfo
- Publication number
- CN114357126A CN114357126A CN202011090931.7A CN202011090931A CN114357126A CN 114357126 A CN114357126 A CN 114357126A CN 202011090931 A CN202011090931 A CN 202011090931A CN 114357126 A CN114357126 A CN 114357126A
- Authority
- CN
- China
- Prior art keywords
- user
- question
- intelligent
- quality inspection
- module
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an intelligent question-answering system which comprises a user portrait module, an intelligent recommendation module, an answer search module and an intelligent quality inspection module, wherein the user portrait module is used for determining user portrait information of a user from a user portrait library in a conversation scene and displaying the user portrait information on a display interface of the intelligent question-answering system, the intelligent recommendation module is used for identifying the intention of a first question proposed by the user and determining an answer corresponding to the intention of the first question according to a question-answering knowledge base, the answer search module is used for identifying the intention of a second question input by an agent and determining an answer corresponding to the second question according to the question-answering knowledge base, and the intelligent quality inspection module is used for performing quality inspection on a conversation text of the agent according to preset quality inspection rules and feeding back a quality inspection result to the agent. The method can automatically, accurately and timely assist the operator to operate, thereby solving the problem that the answer recovery efficiency is very low because the long processing time is consumed for manually searching the answer in the prior art.
Description
Technical Field
The invention relates to the field of financial technology (Fintech), in particular to an intelligent question-answering system.
Background
With the development of computer technology, more and more technologies are applied in the financial field, and the traditional financial industry is gradually changing to financial technology, but due to the requirements of the financial industry on safety and real-time performance, higher requirements are also put forward on the technologies. In the customer service in the financial field, it has become a great challenge for customer service to quickly and accurately reply to the problem posed by the customer in the conversation, and if the problem posed by the customer in the conversation can be timely and accurately replied, the method has great significance for improving the whole conversation quality and improving the experience of the customer.
The existing customer service question-answering system mainly replies to questions posed by customers based on the fact that operators are skilled in advance to know business and product knowledge or on the basis of question-answering materials organized in advance. For example, after a client proposes a certain consultation question in a conversation, an agent person needs to search for an answer from materials such as faq (Frequently answered Questions) handbooks, words, excels, books and the like, and then organizes a language based on the searched corresponding answer to send the language as a reply to the consultation question to the client. However, the manual search faq for the materials such as handbook, word, excel, book, etc. takes a long time to process, resulting in low recovery efficiency.
In summary, there is a need for an intelligent question-answering system for solving the problem of low reply efficiency caused by long processing time for searching answers manually in the prior art.
Disclosure of Invention
The invention provides an intelligent question-answering system, which is used for solving the problem that the answering efficiency is low due to the fact that long processing time is consumed for manually searching answers in the prior art.
In a first aspect, the invention provides an intelligent question-answering system, which comprises a user portrait module, an intelligent recommendation module, an answer search module and an intelligent quality inspection module;
the user portrait module is used for determining user portrait information of a user from a user portrait library in a conversation scene and displaying the user portrait information on a display interface of the intelligent question-answering system;
the intelligent recommendation module is used for identifying the intention of a first question according to the first question proposed by the user in the conversation scene, determining an answer corresponding to the intention of the first question according to a question-answer knowledge base, and displaying the answer corresponding to the intention of the first question on a display interface of the intelligent question-answer system;
the answer searching module is used for identifying the intention of a second question according to the second question input by the seat personnel in the conversation scene, determining an answer corresponding to the intention of the second question according to the question-answer knowledge base, and displaying the answer corresponding to the intention of the second question on a display interface of the intelligent question-answer system;
and the intelligent quality inspection module is used for performing quality inspection on the dialog text of the seat personnel in the dialog scene according to a preset quality inspection rule and feeding back a quality inspection result to the seat personnel.
In the technical scheme, the user portrait module in the intelligent question-answering system displays user portrait information of the user on a display interface of the intelligent question-answering system in real time when an agent and the user have a conversation, and the displayed user portrait information is richer and more timely, so that the agent can be released from heavy labor for manually arranging the user information; the intelligent recommendation module can analyze a first question provided by a user in real time when the seat personnel converse with the user, determines an answer corresponding to the intention of the first question from a question-answer knowledge base according to the intention of the first question provided by the user, and displays the answer corresponding to the intention of the first question on a display interface of an intelligent question-answer system for the seat personnel to reply by reference, so that the reply accuracy and efficiency of the seat personnel can be greatly improved, and the experience of the user is improved; when the seat personnel want to search for a second question, the answer searching module identifies the second question which the seat personnel want to search for, determines the intention of the second question, determines the answer corresponding to the intention of the second question according to the question-answer knowledge base, and displays the answer corresponding to the intention of the second question on the display interface of the intelligent question-answer system so as to assist the seat personnel to perform corresponding processing, thereby greatly improving the processing efficiency of the seat personnel; when the agent personnel talk with the user, the intelligent quality inspection module performs quality inspection on the dialogue text of the agent personnel in the dialogue according to a preset quality inspection rule and feeds back the quality inspection result to the agent personnel so that the agent personnel can standardize the agent behavior according to the quality inspection result. Therefore, the intelligent question-answering system integrates the user image module, the intelligent recommending module and the intelligent quality inspection module on one display interface, can automatically, accurately and timely assist the seat personnel to operate, and is beneficial to reducing the working strength of the seat personnel, so that the problem that in the prior art, the response efficiency is very low due to the fact that long processing time is consumed for manually searching answers can be solved.
Optionally, the user image module obtains the ID of the user or the user query request of the seat staff in the dialog scene, and determines the user image information of the user corresponding to the ID of the user or the user query request of the seat staff from a user image library according to the ID of the user or the user query request of the seat staff.
Optionally, the user portrait module compares the ID of the user with user portrait information in the user portrait library to determine user portrait information of the user corresponding to the ID of the user.
Optionally, the user portrait module performs semantic recognition on the user query request, determines a query intention of the agent staff, performs similarity matching between the query intention of the agent staff and the user portrait information in the user portrait library, and determines the user portrait information of the user corresponding to the query intention of the agent staff.
In the technical scheme, when the user portrait module is in conversation with the user, the user portrait module can determine the user portrait information corresponding to the user ID or the user query information input by the agent from the user portrait library through the ID of the user or the intention of the user query information input by the agent, so that the user portrait module can accurately display the user portrait information of the user on the display interface of the intelligent question and answer system in time, the agent can be released from the heavy labor of manually arranging the user information, and the corresponding operation of the agent according to the user portrait information of the user can be facilitated.
Optionally, the user portrait module obtains the historical data of the user, analyzes and processes the historical data of the user, establishes user portrait information of the user according to the analyzed and processed historical data of the user, and stores the user portrait information of the user in the user portrait library.
According to the technical scheme, the historical data of the user is analyzed, the user portrait information of the user is established according to the analyzed and processed historical data of the user, and the user portrait information of the user is stored in the user portrait library, so that the intelligent question answering system can be beneficial to enabling an agent person to timely and accurately display the user portrait information of the user on a display interface of the intelligent question answering system when the agent person has a conversation with the user, and the agent person can perform corresponding operation processing according to the user portrait information of the user.
Optionally, the user representation module is further configured to query historical conversation contents of the user and the agent person in the conversation scene.
In the technical scheme, the agent personnel can inquire the historical conversation content of the user in the user portrait module, and the agent personnel can inquire corresponding useful information from the historical conversation content according to the requirements of the agent personnel so as to be referred by the agent personnel.
Optionally, the intelligent recommendation module performs semantic recognition on the first question in combination with the context content of the first question raised by the user, and determines the intention of the first question; matching the intention of the first question with the answers in the question-answer knowledge base, determining a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the matching degrees with the sequence larger than a preset threshold value on a display interface of the intelligent question-answer system so that the seat personnel can select the answers as answers for replying the user.
Among the above-mentioned technical scheme, the intelligence recommendation module carries out semantic recognition to this first problem in combination with the context of the first problem that the user proposed, determine the intention of first problem, match the intention of first problem and the answer in the knowledge base of asking for answering again, determine a plurality of matching degrees, later show the answer that the matching degree that ranks before corresponds on the display interface of intelligence asking for answering system, reply with reference to for the seat personnel, can help improving seat personnel's reply precision and efficiency greatly, thereby promote user's experience.
Optionally, the answer searching module performs semantic recognition on a second question input by the agent personnel to determine the intention of the second question; matching the intention of the second question with the answers in the question-answer knowledge base to determine a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the sequenced matching degrees on a display interface of the intelligent question-answer system for the reference of the seat personnel.
Among the above-mentioned technical scheme, answer search module carries out semantic recognition to the second problem of seat personnel input, determines the intention of second problem to match the intention of this second problem with the answer in the knowledge base of asking for answering, determine a plurality of matching degree, show the answer that a plurality of matching degree correspond on intelligent asking for answering system's display interface again, so that supplementary seat personnel carry out corresponding processing, greatly promoted seat personnel's treatment effeciency.
Optionally, the intelligent quality inspection module obtains a dialog text of an agent in the dialog scene, performs quality inspection on the dialog text of the agent according to each quality inspection item in the preset quality inspection rule, determines a quality inspection score corresponding to each quality inspection item in the preset quality inspection rule, and feeds back the quality inspection score corresponding to each quality inspection item in the preset quality inspection rule to the agent so as to prompt the attention of the agent in the dialog scene.
According to the technical scheme, the intelligent quality inspection module performs real-time quality inspection on the dialog text of the seat personnel according to each quality inspection item in the preset quality inspection rule, and feeds back the quality inspection score corresponding to each quality inspection item in the preset quality inspection rule to the seat personnel, so that the attention items of the seat personnel in the dialog can be prompted in time, the seat behaviors of the seat personnel are normalized, and the operation enthusiasm of the seat personnel is improved.
Optionally, the intelligent quality inspection module obtains a conversation record of an agent in the conversation scene, performs voice recognition on the conversation record of the agent, and converts the conversation voice of the agent into a conversation text corresponding to the conversation record of the agent.
Among the above-mentioned technical scheme, intelligent quality control module carries out speech recognition to the dialogue recording of agent personnel, converts agent personnel's dialogue pronunciation into the dialogue text that agent personnel's dialogue recording corresponds to intelligent quality control module can carry out real-time quality control according to the dialogue text of preset quality control rule to agent personnel.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic diagram of an intelligent question answering system architecture according to an embodiment of the present invention;
fig. 2 is a schematic functional interface diagram of an intelligent question answering system in a customer telephone complaint scenario according to an embodiment of the present invention;
FIG. 3 is a flow diagram of a user representation module according to an embodiment of the present invention;
fig. 4 is a schematic flowchart of an intelligent recommendation module according to an embodiment of the present invention;
fig. 5 is a schematic flowchart of an answer search module according to an embodiment of the present invention;
fig. 6 is a schematic flow chart of an intelligent quality inspection module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is an architecture of an intelligent question answering system according to an embodiment of the present invention. As shown in FIG. 1, the intelligent question-answering system may include a user representation module 101, an intelligent recommendation module 102, an answer search module 103, and an intelligent quality inspection module 104. The intelligent question-answering system integrates tools which are required to search knowledge and reply users in a plurality of places into one system interface, and assists the operators to operate according to one-key and simple operation requirements, so that the working intensity of the operators can be greatly reduced, the operating efficiency of the operators is greatly improved, and the user experience of customer consultation is improved. In addition, the intelligent question-answering system can be applied to not only voice customer channels (such as telephone voice, online audio stream and the like) but also text customer channels (such as a mobile phone APP terminal, a tablet computer APP terminal, a webpage h5 terminal and the like).
The user portrait module 101 is configured to determine user portrait information of a user from a user portrait library in a dialog scene and display the user portrait information on a display interface of the intelligent question answering system.
Specifically, the user image module 101 obtains an ID of a user or a user query request of an agent in a conversation scene, compares the ID of the user with user image information in a user image library, and determines user image information of the user corresponding to the ID of the user; or performing semantic recognition on the user query request to determine the query intention of the seat personnel, performing similarity matching on the query intention of the seat personnel and the user image information in the user image library, and determining the user image information of the user corresponding to the query intention of the seat personnel.
For example, when an agent recommends a financial product to a customer a, the user portrait module 101 may obtain an ID (identification number) of the customer a, extract corresponding user portrait information of the customer a (such as name, age, gender, academic calendar, hobbies, occupation, income, credit, family member composition, city of residence, and the like of the customer a) from the user portrait library according to the ID of the customer a, and timely and accurately display the user portrait information of the customer a on a display interface of the intelligent question and answer system, so that the agent may accurately recommend a financial product to the customer a according to the user portrait information of the customer a, thereby quickly completing an order. Or the user portrait module 101 inputs the user portrait information of Zhang III of relatives of the inquiry client A, the user portrait module 101 identifies the intention of the user portrait information of Zhang III that the agent wants to inquire by the agent through the identification of the inquiry request of the agent, and then extracts the corresponding user portrait information of Zhang III from the user portrait library according to the association relationship between the client A and Zhang III, so that the agent can recommend corresponding financial products to Zhang III according to the user portrait information of Zhang III.
It should be noted that, in the existing customer service question and answer system, it is difficult to perform operations accurately because the operators have little knowledge about the customers and are not in time. For example, in a telephone sales scene, if the staff of the seat does not have a certain control on the work, income, credit investigation, family member composition, etc. of the customer, it is difficult to find the pain point of the user, and it is difficult to quickly reach the order and avoid the business risk. Therefore, the user profile module 101 can automatically analyze business system data, central credit data, web crawler data, and other data in data sets, and build a comprehensive user profile for an entity (corporate, natural, etc.), including basic attribute information, analysis tags, etc. of the entity. When the seat personnel converse with the user, the user image information of the user can be displayed on the conversation page in real time, so that the seat personnel can be released from heavy labor of manually arranging the user information, and the user information displayed by the user image module 101 is richer and more timely.
The intelligent recommendation module 102 is configured to identify an intention of a first question according to the first question posed by the user in the dialog scene, determine an answer corresponding to the intention of the first question according to the question-answer knowledge base, and display the answer corresponding to the intention of the first question on a display interface of the intelligent question-answer system.
Specifically, the intelligent recommendation module 102 performs semantic recognition on the first question in combination with the context of the first question presented by the user, and determines the intention of the first question. And matching the intention of the first question with the answers in the question-answer knowledge base to determine a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the matching degrees with the sequence greater than a preset threshold value on a display interface of the intelligent question-answer system so as to be selected by the seat personnel as the answers of the reply user. Wherein the preset threshold value can be set empirically.
Illustratively, taking a customer B telephone complaint scene as an example for description, as shown in fig. 2, when an agent has a conversation with a customer B, user portrait information of the customer B is displayed on a display interface of an intelligent question-answering system in real time, the customer B says "i want to complain about your staff and affect my life and too much people" in the telephone, meanwhile, ASR (Automatic Speech Recognition) performs real-time Speech Recognition on Speech in the customer B telephone, converts the Speech in the customer B telephone into text content, and displays the text content on the display interface of the intelligent question-answering system, the intelligent recommendation module 102 performs real-time analysis and Recognition on the question proposed by the customer B, recognizes the intention "complaint staff" of the customer B, finds out a plurality of relevant answers from a question-answering knowledge base according to the intention of the customer B, and sorting the searched multiple related answers according to the relevance size of the answers, and displaying the most related answers (such as' do you get good, regress the inconvenience brought to you deeply,.
It should be noted that, since the business required to be contacted is too much knowledge for the human agent, it is difficult to be skillfully grasped in a short period of time. For example, in a banking scene, services include loan, deposit, financing and the like, each service includes several to hundreds of products, and it is difficult for an attendant to master the knowledge of all services and products in a short time, which results in that after a user asks questions, the attendant needs to find answers from faq manuals, words, excels, books and other materials. Then, the language is organized according to the found answer and sent to the user as a reply, so that the reply efficiency is low. Therefore, the intelligent recommendation module 102 can analyze the questions posed by the user in real time, search the relevant answers from the knowledge base, sort the answers according to the relevance of the answers from large to small, and display the most relevant answers on the display interface of the intelligent question-answering system for the response of the seat personnel, so that the response precision and efficiency of the seat personnel are greatly improved, and the user experience is facilitated to be improved.
The answer search module 103 is configured to identify an intention of a second question according to the second question input by the agent in the dialog scene, determine an answer corresponding to the intention of the second question according to the question-answer knowledge base, and display the answer corresponding to the intention of the second question on a display interface of the intelligent question-answer system.
Specifically, the answer searching module 103 performs semantic recognition on the second question input by the seat person, and determines the intention of the second question. And matching the intention of the second question with the answers in the question-answer knowledge base to determine a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the sequenced matching degrees on a display interface of the intelligent question-answer system for the reference of the seat personnel.
Continuing with the example of a customer B telephone complaint scenario, as shown in fig. 2, during the course of the seat person's conversation with customer B, the seat person knows that customer B is because the loan is not due, the customer B loan is not informed of his wife, and the acquirer calls his wife again to customer B to make a call to him to make his debt, so that customer B complaints the clerk. In view of this, the agent person inputs the question "overdue collection prompt attention specification" to be queried in the answer search module 103, the answer search module 103 performs semantic recognition on the question queried by the agent person, determines the intention "query overdue collection prompt attention item" of the agent person, searches a plurality of relevant answers to the question from the question-answer knowledge base according to the intention, sorts the searched plurality of relevant answers according to the relevance of the answers, and then displays the sorted plurality of relevant answers (such as "the acquirer must not ask for money, hurt, illegal obliged debtor … …", "the acquirer must not induce or compel the debtor to collect the fund collection debt through new loan or illegal approach", "must not ask for collection of the part beyond the national relevant regulation law" and the like) on the display interface of the intelligent question-answer system, for the reference of the seat personnel. In addition, the answer search module 103 supports modification of the content of the dialogs in addition to searching the knowledge base of questions and answers and the query dialogs.
The intelligent quality inspection module 104 is used for performing quality inspection on the dialog text of the seat personnel in the dialog scene according to a preset quality inspection rule and feeding back a quality inspection result to the seat personnel.
Specifically, the intelligent quality inspection module 104 obtains the conversation sound of the seat person in the conversation scene, performs voice recognition on the conversation sound of the seat person, and converts the conversation voice of the seat person into a conversation text corresponding to the conversation sound of the seat person. And then, performing quality inspection on the dialog text of the seat personnel according to each quality inspection item in the preset quality inspection rule, determining a quality inspection score corresponding to each quality inspection item in the preset quality inspection rule, and feeding back the quality inspection score corresponding to each quality inspection item in the preset quality inspection rule to the seat personnel so as to prompt the attention of the seat personnel in a dialog scene.
Continuing with the example of a customer B telephone complaint scenario, as shown in fig. 2, during the course of the seat person's conversation with customer B, the seat person knows that customer B is because the loan is not due, the customer B loan is not informed of his wife, and the acquirer calls his wife again to customer B to make a call to him to make his debt, so that customer B complaints the clerk. The intelligent quality inspection module 104 performs real-time quality inspection on the dialog text of the agent person according to pre-configured quality inspection item requirements (such as the requirement of violent quality inspection items, etc.), such as the detection of sensitive words (such as the agent person abuse client B "impounding", the agent person abuse client B "tmd", the client B emotional activation, etc.), and the detected sensitive words are scored, the quality inspection scoring result is fed back to the seat personnel, and the attention in the conversation of the seat personnel is prompted in time (for example, "no alarm" for the customer B suffering from abuse, prompt "please regress for errors immediately", tmd "for the customer B suffering from abuse, prompt" regress for errors immediately ", prompt" please regress for emotions of the customer B ", and the like), so that the monitoring of the seat behavior of the seat personnel can be facilitated. In addition, the intelligent quality control module 104 may use a pop-up bubble to alert the attendant to a notice in the dialog.
It should be noted that, the existing seat system cannot monitor the service quality of the seat personnel in time, which causes dissatisfaction of customers and service risk. There are some dialects in the seat service process that must be said otherwise, business risks may be caused, such as the need to inform the loan in the loan electric check that the loan cannot be used for house purchase, stock handling, etc. Some dialogs in the agent process, such as the occurrence of an expletor, complaint, etc. related content in the dialog, may cause user dissatisfaction once they are misspoken. Therefore, the intelligent quality inspection module 104 can perform real-time quality inspection on the dialog text of the seat personnel according to the pre-configured quality inspection item requirements, feed back the quality inspection result to the seat personnel, prompt the attention items in the dialog of the seat personnel in time and supervise and standardize the seat behavior.
In addition, as shown in the service flow page of fig. 2, the service flow node of the seat staff is automatically checked after the completion, and if there is a missing flow link, the seat staff is reminded in time. And the seat personnel can also automatically switch scenes according to the conversation content with the user. As shown in the real-time speech recognition page of fig. 2, the intelligent question and answer system highlights some key information in the dialog process between the agent and the user, and also supports the copy of the dialog information in the dialog process between the agent and the user.
It should be noted that the system structure shown in fig. 1 is only an example, and the embodiment of the present invention is not limited thereto.
Based on the above description, the flow of the user profile module of the present invention is described below through a specific implementation scenario in order to better explain embodiments of the present invention.
As shown in fig. 3, the process includes the following steps:
In the embodiment of the invention, when the seat personnel converse with the user, the ID of the user or the user query request of the seat personnel can be obtained.
In the embodiment of the invention, the semantic recognition is carried out on the user query request of the seat personnel to determine the query intention of the seat personnel.
In the embodiment of the invention, the ID of the user is compared with the user portrait information in the user portrait library to determine the user portrait information of the user corresponding to the ID of the user; or carrying out similarity matching on the inquiry intention of the seat personnel and the user image information in the user image library to determine the user image information of the user corresponding to the inquiry intention of the seat personnel.
The embodiment shows that when the user portrait module is in conversation with the user, the user portrait module can determine the user portrait information corresponding to the user ID or the user query information input by the agent from the user portrait library through the ID of the user or the intention of the user query information input by the agent, so that the user portrait module can accurately display the user portrait information of the user on the display interface of the intelligent question and answer system in time, and the displayed user portrait information is richer and more timely, so that the agent can be released from heavy labor for manually arranging the user information, and the user portrait module is favorable for the agent to perform corresponding operation according to the user portrait information of the user.
In order to better explain the embodiment of the present invention, the flow of the intelligent recommendation module of the present invention is described below through a specific implementation scenario.
As shown in fig. 4, the process includes the following steps:
In the embodiment of the invention, the problem proposed by the user in the conversation scene and the context content corresponding to the problem are obtained.
In the embodiment of the invention, the semantic recognition is carried out on the question by combining the context content of the question proposed by the user, and the intention of the question is determined.
In step 403, a matching module is searched.
In the embodiment of the invention, the intention of the question is matched with the answers in the question-answer knowledge base, and a plurality of matching degrees are determined.
In the embodiment of the invention, the matching degrees are sorted according to the matching degrees, and the answers corresponding to the matching degrees with the sorted order larger than the preset threshold value are displayed on the display interface of the intelligent question-answering system so as to be selected by the seat staff as the answers of the reply user.
The above embodiment shows that the intelligent recommendation module performs semantic recognition on the question by combining the context of the question provided by the user, determines the intention of the question, matches the intention of the question with the answers in the question-answer knowledge base, determines a plurality of matching degrees, and then displays the answer corresponding to the matching degree before ranking on the display interface of the intelligent question-answer system for the seat personnel to reply with reference, which can help to greatly improve the reply accuracy and efficiency of the seat personnel, thereby improving the experience of the user.
In order to better explain the embodiment of the present invention, the flow of the answer searching module of the present invention is described below through a specific implementation scenario.
As shown in fig. 5, the process includes the following steps:
In the embodiment of the invention, the problem of the input of the seat personnel in the conversation scene is obtained.
In the embodiment of the invention, the semantic recognition is carried out on the problem input by the seat personnel to determine the intention of the problem.
In the embodiment of the invention, the intention of the question is matched with the answers in the question-answer knowledge base, and a plurality of matching degrees are determined.
And 505, returning the result set to the agent personnel.
In the embodiment of the invention, the matching degrees are sorted according to the matching degrees, and answers corresponding to the sorted matching degrees are displayed on a display interface of the intelligent question-answering system for the reference of an agent.
The above embodiment shows that the answer search module performs semantic recognition on the questions input by the seat personnel, determines the intention of the questions, matches the intention of the questions with the answers in the question and answer knowledge base, determines a plurality of matching degrees, and displays the answers corresponding to the matching degrees on the display interface of the intelligent question and answer system, so as to assist the seat personnel to perform corresponding processing, and greatly improve the processing efficiency of the seat personnel.
In order to better explain the embodiment of the present invention, the flow of the intelligent quality inspection module of the present invention is described below through a specific implementation scenario.
As shown in fig. 6, the process includes the following steps:
In the embodiment of the invention, the pre-configured quality inspection item requirement is obtained.
And step 603, scoring the quality inspection items.
In the embodiment of the invention, the dialog text of the seat personnel is subjected to quality inspection according to the pre-configured quality inspection item requirements, and the quality inspection score corresponding to each quality inspection item requirement in the pre-configured quality inspection item requirements is determined.
In the embodiment of the invention, the quality inspection score corresponding to each quality inspection item in the pre-configured quality inspection item requirements is counted, and the overall quality inspection score of the seat personnel is counted when the conversation is finished. And then feeding back the quality inspection score corresponding to each quality inspection item requirement in the pre-configured quality inspection item requirements and the overall quality inspection score of the seat personnel to the seat personnel so as to prompt the attention of the seat personnel in a conversation scene.
The embodiment shows that the intelligent quality inspection module performs real-time quality inspection on the dialog text of the seat personnel according to each quality inspection item in the preset quality inspection rule, and feeds back the quality inspection score corresponding to each quality inspection item in the preset quality inspection rule to the seat personnel, so that the intelligent quality inspection module can help prompt the attention items of the seat personnel in the dialog in time, standardizes the seat behaviors of the seat personnel, and improves the operation enthusiasm of the seat personnel.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present application and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An intelligent question-answering system is characterized by comprising a user portrait module, an intelligent recommendation module, an answer search module and an intelligent quality inspection module;
the user portrait module is used for determining user portrait information of a user from a user portrait library in a conversation scene and displaying the user portrait information on a display interface of the intelligent question-answering system;
the intelligent recommendation module is used for identifying the intention of a first question according to the first question proposed by the user in the conversation scene, determining an answer corresponding to the intention of the first question according to a question-answer knowledge base, and displaying the answer corresponding to the intention of the first question on a display interface of the intelligent question-answer system;
the answer searching module is used for identifying the intention of a second question according to the second question input by the seat personnel in the conversation scene, determining an answer corresponding to the intention of the second question according to the question-answer knowledge base, and displaying the answer corresponding to the intention of the second question on a display interface of the intelligent question-answer system;
and the intelligent quality inspection module is used for performing quality inspection on the dialog text of the seat personnel in the dialog scene according to a preset quality inspection rule and feeding back a quality inspection result to the seat personnel.
2. The intelligent question answering system according to claim 1, wherein the user imaging module obtains the user ID of the user or the user query request of the seat staff in the conversation scene, and determines the user image information of the user corresponding to the user ID of the user or the user query request of the seat staff from a user imaging library according to the user ID of the user or the user query request of the seat staff.
3. The intelligent question answering system of claim 2 wherein the user representation module compares the user ID with user representation information in the user representation library to determine user representation information for the user corresponding to the user ID.
4. The intelligent question answering system according to claim 2, wherein the user portrait module performs semantic recognition on the user query request to determine a query intention of the agent person, performs similarity matching between the query intention of the agent person and user portrait information in the user portrait library, and determines user portrait information of the user corresponding to the query intention of the agent person.
5. The intelligent question answering system according to claim 2, wherein the user representation module obtains historical data of the user, analyzes the historical data of the user, establishes user representation information of the user according to the analyzed historical data of the user, and stores the user representation information of the user in the user representation library.
6. The intelligent question-answering system of claim 1, wherein the user profile module is further configured to query historical conversation content of the user and the agent person in the conversation scenario.
7. The intelligent question answering system according to claim 1, wherein the intelligent recommendation module semantically identifies a first question posed by the user in combination with context of the first question to determine an intent of the first question; matching the intention of the first question with the answers in the question-answer knowledge base, determining a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the matching degrees with the sequence larger than a preset threshold value on a display interface of the intelligent question-answer system so that the seat personnel can select the answers as answers for replying the user.
8. The intelligent question-answering system according to claim 1, wherein the answer searching module performs semantic recognition on a second question input by the agent person to determine the intention of the second question; matching the intention of the second question with the answers in the question-answer knowledge base to determine a plurality of matching degrees, sequencing the matching degrees according to the matching degrees, and displaying the answers corresponding to the sequenced matching degrees on a display interface of the intelligent question-answer system for the reference of the seat personnel.
9. The intelligent question-answering system according to claim 1, wherein the intelligent quality inspection module acquires a dialog text of an agent in the dialog scene, performs quality inspection on the dialog text of the agent according to each quality inspection item in the preset quality inspection rule, determines a quality inspection score corresponding to each quality inspection item in the preset quality inspection rule, and feeds back the quality inspection score corresponding to each quality inspection item in the preset quality inspection rule to the agent to prompt the attention of the agent in the dialog scene.
10. The intelligent question-answering system according to claim 9, wherein the intelligent quality inspection module obtains a conversation voice of an agent in the conversation scene, performs voice recognition on the conversation voice of the agent, and converts the conversation voice of the agent into a conversation text corresponding to the conversation voice of the agent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011090931.7A CN114357126A (en) | 2020-10-13 | 2020-10-13 | Intelligent question-answering system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011090931.7A CN114357126A (en) | 2020-10-13 | 2020-10-13 | Intelligent question-answering system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114357126A true CN114357126A (en) | 2022-04-15 |
Family
ID=81090061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011090931.7A Pending CN114357126A (en) | 2020-10-13 | 2020-10-13 | Intelligent question-answering system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114357126A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116578682A (en) * | 2023-05-22 | 2023-08-11 | 浙江法之道信息技术有限公司 | Intelligent consultation method and system for legal service |
| CN116739235A (en) * | 2023-05-06 | 2023-09-12 | 广州圈量网络信息科技有限公司 | Intelligent community service management system based on big data |
| CN117408708A (en) * | 2023-11-09 | 2024-01-16 | 南方电网储能股份有限公司信息通信分公司 | Customer service center dispatching system based on big data |
-
2020
- 2020-10-13 CN CN202011090931.7A patent/CN114357126A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116739235A (en) * | 2023-05-06 | 2023-09-12 | 广州圈量网络信息科技有限公司 | Intelligent community service management system based on big data |
| CN116739235B (en) * | 2023-05-06 | 2024-03-22 | 广州圈量网络信息科技有限公司 | Intelligent community service management system based on big data |
| CN116578682A (en) * | 2023-05-22 | 2023-08-11 | 浙江法之道信息技术有限公司 | Intelligent consultation method and system for legal service |
| CN116578682B (en) * | 2023-05-22 | 2024-02-13 | 浙江法之道信息技术有限公司 | Intelligent consultation method and system for legal service |
| CN117408708A (en) * | 2023-11-09 | 2024-01-16 | 南方电网储能股份有限公司信息通信分公司 | Customer service center dispatching system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109767787B (en) | Emotion recognition method, device and readable storage medium | |
| CN112804400B (en) | Customer service call voice quality inspection method and device, electronic equipment and storage medium | |
| CN114357126A (en) | Intelligent question-answering system | |
| US7711566B1 (en) | Systems and methods for monitoring speech data labelers | |
| US9582757B1 (en) | Scalable curation system | |
| US20100332287A1 (en) | System and method for real-time prediction of customer satisfaction | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| KR102523638B1 (en) | Method of real time intent recognition | |
| EP4036823A1 (en) | Customization platform and method for service quality evaluation product | |
| JP2017215931A (en) | Conference support system, conference support device, conference support method, and program | |
| US10192569B1 (en) | Informing a support agent of a paralinguistic emotion signature of a user | |
| WO2021136458A1 (en) | Processing method and apparatus for police emergency dispatching | |
| CN112084318B (en) | Dialogue assistance method, system and device | |
| US10255346B2 (en) | Tagging relations with N-best | |
| CN110457454A (en) | A kind of dialogue method, server, conversational system and storage medium | |
| CN114911929A (en) | Classification model training method, text mining equipment and storage medium | |
| KR20110048675A (en) | Call center counsel method and counsel system using voice recognition and tagging | |
| JP2020160425A (en) | Evaluation system, evaluation method, and computer program | |
| CN111651554A (en) | Insurance question-answer method and device based on natural language understanding and processing | |
| CN115564529A (en) | Voice navigation control method and device, computer terminal and storage medium | |
| CN113505606B (en) | Training information acquisition method and device, electronic equipment and storage medium | |
| CN110765242A (en) | Method, device and system for providing customer service information | |
| CN113420140B (en) | User emotion prediction method and device, electronic equipment and readable storage medium | |
| US11837251B2 (en) | Virtual counseling system and counseling method using the same | |
| CN114186048A (en) | Question-answer replying method and device based on artificial intelligence, computer equipment and 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 |