Method and related media for full-cycle automated intelligent customer service handling

Abstract

This application discloses a fully automated intelligent customer service system, related methods, and related media based on converged communications. The system includes: a security management subsystem, used to verify whether each converged communication user meets the login conditions, and allowing each converged communication user to log in to the joint center subsystem if the login conditions are met; a joint center subsystem, used to determine a target node device among multiple node devices to perform intelligent question answering; instructing the target node device to receive each inquiry information sent by each converged communication user and generate a first feedback information corresponding to each inquiry information; and sending each inquiry information to the intelligent processing subsystem when the target node device does not generate the first feedback information. The solution obtained by this application can effectively improve the intelligence level of the converged communication user inquiry process and improve the security and efficiency of problem solving.

Description

Technical Field

[0001] This application relates to the field of converged communication technology, and in particular to a fully automated intelligent customer service system based on converged communication, related methods, and related media. Background Technology

[0002] With the development of information technology, the era of 5G (5th generation wireless systems) is gradually arriving. The application of 5G technology has brought a qualitative leap to the field of communications, greatly promoting the development of converged communication technology. Converged communication is a new communication model that uses IP communication as its foundation and integrates computer technology with traditional communication technologies to establish a network platform that includes both computer networks and traditional communication networks. It integrates numerous application services, including instant messaging, Voice over Internet Protocol (VoIP) telephony, video communication, data transmission, multimedia conferencing, and collaborative office work. The application areas of converged communication are becoming increasingly diversified.

[0003] Currently, although customer service systems are transitioning towards intelligence, the overall level of intelligence is low. The chatbot's question-and-answer function is too mechanical, only able to answer a very small number of simple, programmed questions. Slightly more complex or differentiated questions still rely on human customer service representatives. This results in a slow and inefficient question-and-answer process. Summary of the Invention

[0004] This application provides a fully automated intelligent customer service system based on converged communications, which can effectively improve the intelligence level of the user inquiry process and enhance the security and efficiency of problem solving.

[0005] On the one hand, a fully automated intelligent customer service system based on converged communications is characterized in that the system includes:

[0006] The security management subsystem is used to verify whether each of the multiple converged communication users meets the login conditions, and if each converged communication user meets the login conditions, allow each converged communication user to log in to the joint center subsystem;

[0007] The joint center subsystem includes multiple node devices, which are used to determine the target node device among the multiple node devices for intelligent question answering.

[0008] The joint center subsystem is also used to instruct the target node device to receive each query message sent by each converged communication user, and to generate a first feedback message corresponding to each query message;

[0009] The joint center subsystem is also used to send each query information to the intelligent processing subsystem when the target node device has not generated the first feedback information;

[0010] The intelligent processing subsystem is used to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0011] On one hand, an intelligent customer service method is applied to a fully automated intelligent customer service system based on converged communications. The system includes a security management subsystem, a joint center subsystem, and an intelligent processing subsystem. The method includes:

[0012] The security management subsystem is used to verify whether each of the multiple converged communication users meets the login conditions, and if each converged communication user meets the login conditions, allow each converged communication user to log in to the joint center subsystem;

[0013] The joint center subsystem includes multiple node devices, which are used to determine the target node device among the multiple node devices for intelligent question answering.

[0014] The joint center subsystem is also used to instruct the target node device to receive each query message sent by each converged communication user, and to generate a first feedback message corresponding to each query message;

[0015] The joint center subsystem is also used to send each query information to the intelligent processing subsystem when the target node device has not generated the first feedback information;

[0016] The intelligent processing subsystem is used to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0017] On one hand, an intelligent customer service device is applied to a fully automated intelligent customer service system based on converged communications, the device comprising:

[0018] The security management unit is used to verify whether each of the multiple converged communication users meets the login conditions, and if each converged communication user meets the login conditions, allow each converged communication user to log in to the joint center subsystem;

[0019] The joint central unit includes multiple node devices, which are used to determine the target node device among the multiple node devices for intelligent question answering.

[0020] The joint central unit is also used to instruct the target node device to receive each query message sent by each converged communication user, and to generate a first feedback message corresponding to each query message;

[0021] The joint central unit is also used to send each query information to the intelligent processing subsystem when the target node device has not generated the first feedback information;

[0022] The intelligent processing unit is configured to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0023] On one hand, embodiments of this application provide an electronic device, including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing the steps in the aforementioned fully automated intelligent customer service system based on converged communication.

[0024] Accordingly, embodiments of this application provide a computer-readable storage medium for storing computer program instructions for use by a terminal device, which includes programs involved in executing steps in a fully automated intelligent customer service system based on converged communications.

[0025] Accordingly, embodiments of this application provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. When the computer instructions are executed by the processor of a computer device, they perform the methods described in the above embodiments. This computer program product can be a software installation package.

[0026] As can be seen from the embodiments of this application, the fully automated intelligent customer service system based on converged communication comprises various sub-modules that handle different stages of the entire customer service process. The combination of distributed and collaborative processing helps to reduce the information processing load of any particular sub-module and improve the overall system's service efficiency. Furthermore, by integrating the communication technology of the joint center subsystem with the data processing technology of the security management subsystem and the intelligent processing subsystem, converged communication is achieved, which also enhances the system's intelligence and automation levels. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic diagram of the structure of a fully automated intelligent customer service system based on converged communication, provided in an embodiment of this application.

[0029] Figure 2 This is a schematic diagram of the method flow corresponding to a fully automated intelligent customer service system based on converged communication provided in an embodiment of this application;

[0030] Figure 3 This is a schematic diagram of a method flow corresponding to another fully automated intelligent customer service system based on converged communication provided in an embodiment of this application;

[0031] Figure 4 This is a functional unit diagram of a fully automated intelligent customer service device based on converged communication provided in an embodiment of this application;

[0032] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0033] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0034] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0035] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0036] Currently, although customer service systems are transitioning towards intelligence, the overall level of intelligence is low. The chatbot's question-and-answer function is too mechanical, only able to answer a very small number of simple, programmed questions. Slightly more complex or differentiated questions still rely on human customer service representatives. This results in a slow and inefficient question-and-answer process.

[0037] To address the aforementioned issues, this application provides a fully automated intelligent customer service system based on converged communications. A detailed description is provided below with reference to the accompanying drawings.

[0038] First, please refer to Figure 1 The diagram shows a structural schematic of a fully automated intelligent customer service system 100 based on converged communication, which includes a security management subsystem 110, a joint center subsystem 120, and an intelligent processing subsystem 130.

[0039] The aforementioned security management subsystem 110 may include various handheld devices with wireless communication functions (such as smartphones, tablets, etc.), in-vehicle devices, wearable devices (smartwatches, smart bracelets, wireless headphones, augmented reality / virtual reality devices, smart glasses), computing devices or other processing devices connected to a wireless modem, as well as various forms of converged communication user equipment (UE), mobile station (MS), terminal device, etc.

[0040] The aforementioned joint center subsystem includes 120 distributed storage servers, traditional servers, large storage systems, desktop computers, laptops, mobile stations (MS), and terminals.

[0041] The aforementioned intelligent processing subsystem 130 can be a large-scale distributed storage server, a traditional server, a large-scale storage system, a desktop computer, etc.

[0042] When providing customer service to converged communication users, their login information is transmitted to the aforementioned security management subsystem 110. The security management subsystem verifies the login information. Upon successful verification, the user is transferred to the joint center subsystem 120 for a Q&A session. Questions that the joint center subsystem can answer are directly fed back to the converged communication user. For inquiries that cannot be answered, the joint center subsystem forwards the inquiry to the intelligent processing subsystem for further processing and feedback generation, which is then sent back to the converged communication user via the joint center subsystem. This achieves full automation and intelligence throughout the customer service process.

[0043] The technical solutions of the embodiments of this application can be based on Figure 1 The system with the example architecture or its variants is implemented in a concrete way.

[0044] See Figure 2 , Figure 2 This is a flowchart illustrating a fully automated intelligent customer service system based on converged communication, as provided in an embodiment of this application. This method may include, but is not limited to, the following steps:

[0045] 201. Security Management Subsystem, used to verify whether each of the multiple converged communication users meets the login conditions, and if each converged communication user meets the login conditions, allow each converged communication user to log in to the Joint Center Subsystem.

[0046] Specifically, this can be understood as the security management subsystem performing security verification on any one of multiple converged communication users to determine if the user meets the login requirements. This includes verifying the username and password, and may also utilize other verification methods such as facial recognition, iris recognition, or fingerprint recognition. If the converged communication user meets the login requirements, meaning the user verification is successful, the user is allowed to log in to the joint center subsystem, thus gaining access to the query function.

[0047] 202. The joint center subsystem includes multiple node devices, used to determine the target node device among the multiple node devices for intelligent question answering.

[0048] Specifically, in the joint center subsystem comprising multiple node devices, these nodes can be distributed and connected, jointly storing important data, but can execute customer service inquiry operations independently. Each node corresponds to a unique customer service ID for identification. The joint center subsystem determines the target node device for intelligent question answering. This determination can be achieved through real-time status monitoring, identifying currently available node devices awaiting task assignment, and then selecting the target node device from these available devices according to certain criteria or randomly.

[0049] 203. The joint center subsystem is further configured to instruct the target node device to receive each query message sent by each converged communication user, and generate a first feedback message corresponding to each query message.

[0050] Specifically, for any and any query, the joint center subsystem can instruct the target node device to receive the query and generate a first feedback message corresponding to the query. For example, it can search for keywords in the query, retrieve preset feedback messages pre-stored in the database based on the keywords, and then send the preset feedback messages as the first feedback messages to the converged communication user.

[0051] 204. The joint center subsystem is further configured to send each query information to the intelligent processing subsystem when the target node device has not generated the first feedback information.

[0052] Specifically, when the joint center subsystem cannot answer the query information of the converged communication user, such as when the question is too complex or when the preset feedback information is not found, that is, when the first feedback information is not generated, each such query information can be sent to the intelligent processing subsystem, which will intelligently process the query information and generate the corresponding second feedback information.

[0053] 205. The intelligent processing subsystem is used to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0054] Specifically, after the intelligent processing subsystem processes the query information intelligently, it generates corresponding second feedback information. This second feedback information is then sent to the corresponding converged communication user via the joint center subsystem. The specific process by which the intelligent processing subsystem generates the second feedback information is described below.

[0055] As can be seen from the embodiments of this application, the fully automated intelligent customer service system based on converged communication comprises various sub-modules that handle different stages of the entire customer service process. The combination of distributed and collaborative processing helps to reduce the information processing load of a particular sub-module and improve the overall system's service efficiency. Furthermore, by integrating the communication technology of the joint center subsystem with the data processing technology of the security management subsystem and the intelligent processing subsystem, converged communication is achieved, which also enhances the system's intelligence and automation levels.

[0056] With the above Figure 2 The embodiments shown are consistent; please refer to [link / reference]. Figure 3 , Figure 3 This is a flowchart illustrating another intelligent customer service method corresponding to a fully automated intelligent customer service system based on converged communication provided in this application embodiment. The method includes:

[0057] 301. Security management subsystem, used to verify whether each of the multiple converged communication users meets the login conditions, and, if each converged communication user meets the login conditions, allow each converged communication user to log in to the joint center subsystem;

[0058] 302. The joint center subsystem includes multiple node devices, used to determine the target node device among the multiple node devices for intelligent question answering.

[0059] Steps 301-302 above are the same as steps 201-202 mentioned above, and will not be repeated here.

[0060] 303. The joint center subsystem is further configured to extract at least one keyword from each query information to obtain at least one keyword for each query information.

[0061] Specifically, taking text-based inquiry information as an example, when the joint center subsystem extracts keywords from the inquiry information, it can perform semantic analysis to identify content words, function words, and special punctuation marks, thus recognizing the semantics of the inquiry information. Keywords are then selected from the content words based on this semantic understanding. Inquiry information can also be in any other form, such as video, audio, or image. In other forms, image recognition technology and artificial intelligence (AI) technology can be used to first convert the text into text before performing subsequent keyword extraction operations.

[0062] 304. The joint center subsystem is also used to query whether there is preset feedback information for the corresponding query information based on the at least one keyword.

[0063] Specifically, once the joint center subsystem has determined the keywords corresponding to the inquiry information, such as "What is the annual interest rate of XX loan?" and the keywords are "XX loan" and "annual interest rate", it can query whether there is preset feedback information for the corresponding inquiry information based on these keywords. The corresponding preset feedback information could be "The annual interest rate of XX loan is 0.1% of the loan amount", etc.

[0064] 305. The joint center subsystem is further configured to generate first feedback information corresponding to the inquiry information based on the preset feedback information if each inquiry information has preset feedback information.

[0065] Specifically, if the aforementioned preset feedback information exists, then the first feedback information corresponding to the query information can be generated based on the preset feedback information.

[0066] The generation method can either directly use the preset feedback information as the first feedback information, or it can generate a second feedback information based on the preset feedback information.

[0067] 306. The joint center subsystem is further configured to send each query information to the intelligent processing subsystem when the target node device has not generated the first feedback information;

[0068] 307. The intelligent processing subsystem is configured to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0069] Steps 306-307 above are the same as steps 204-205 above, and will not be repeated here.

[0070] As can be seen from the embodiments of this application, the fully automated intelligent customer service system based on converged communication comprises various sub-modules that handle different stages of the entire customer service process. The combination of distributed and collaborative processing helps reduce the information processing load of any single sub-module and improves the overall system's service efficiency. Furthermore, the collaborative center subsystem automatically identifies inquiry information and generates feedback, reducing the need for human customer service and enhancing the intelligence of question-and-answer processing. Simultaneously, by integrating the communication technology of the collaborative center subsystem with the data processing technology of the security management subsystem and the intelligent processing subsystem, converged communication is achieved, which also enhances the system's intelligence and automation levels.

[0071] In one possible example, the security management subsystem is used to verify whether each of a plurality of converged communication users meets the login conditions, including: obtaining the security level of each of the plurality of converged communication users, and determining a target security verification method based on the security level of each converged communication user, wherein the target security verification method includes a first verification method and a second verification method; and verifying whether each converged communication user meets the login conditions based on the target security verification method.

[0072] Specifically, since there are many categories of converged communication users, such as individual converged communication users and enterprise converged communication users, and within each category, different converged communication users can be divided into different levels, each corresponding to a different security level. For example, A is a premium enterprise converged communication user, corresponding to security level 1; B is an ordinary individual converged communication user, corresponding to security level 2; there can also be ordinary enterprise converged communication users C, D, etc. The target security verification method is then determined based on the security level of each converged communication user. For example, security level 1 corresponds to the second verification method, and security level 2 corresponds to the first verification method. Taking converged communication user A and converged communication user B as an example, the security management subsystem uses the second verification method to verify whether converged communication user A meets the login requirements, and uses the first verification method to verify whether converged communication user B meets the login requirements. For other converged communication users with other security levels, third or fourth verification methods can also be used, etc., which will not be listed here.

[0073] It is evident that setting different login verification methods for converged communication users with different security levels is beneficial to ensuring the information security of converged communication users and improving the overall security level of the system.

[0074] In one possible example, verifying whether each converged communication user meets the login conditions according to the target security verification method includes: if the target security verification method of the converged communication user is the first verification method, then verifying the login account information and login password of the converged communication user; and if the login account information and login password of the converged communication user match, determining that the converged communication user has passed the verification.

[0075] Specifically, taking the aforementioned ordinary individual converged communication user B as an example, with a security level of 2 and the corresponding target security verification method being the first verification method, it is only necessary to verify the login account information and login password of converged communication user B; and if the login account information and login password of the converged communication user match, no further verification is required to confirm that the converged communication user B has passed the verification.

[0076] It is evident that for converged communication users with lower security levels, simple verification can streamline the verification process, improve verification efficiency, and facilitate quick login for converged communication users.

[0077] In one possible example, verifying whether each converged communication user meets the login conditions according to the target security verification method includes: if the target security verification method of the converged communication user is the second verification method, then verifying the converged communication user's login account information and login password; and if the converged communication user's login account information and login password match, obtaining a pre-stored public key based on the converged communication user's login account information or login password, generating a standard private key based on the public key using a specific encryption method, and sending the public key to the converged communication user so that the converged communication user generates a private key to be verified based on the public key; receiving the private key to be verified sent by the converged communication user, comparing whether the private key to be verified matches the standard private key, and determining that the converged communication user has passed verification if the private key to be verified matches the standard private key.

[0078] Specifically, taking the aforementioned high-quality enterprise unified communication user A as an example, with a security level of 1, the highest security level, the corresponding target security verification method is the second verification method. This method not only verifies the login account information and password of the unified communication user A, but also, if the login account information and password match, retrieves a pre-stored public key based on the login account information or password. The security management subsystem then generates a standard private key based on the public key using a specific encryption method. The security management subsystem sends the public key to the unified communication user A, enabling the user to generate a private key to be verified. If the private key generated by the unified communication user A uses the same specific encryption method as the private key generated by the security management subsystem, the two are matched, and the verification of the unified communication user A is confirmed.

[0079] The specific encryption method is an encryption algorithm set by the unified communication user during registration. Only the security management subsystem and the unified communication user know the specific encryption method used. The specific encryption method can be an asymmetric encryption algorithm (RSA), a hash function algorithm, etc.

[0080] It is evident that adopting a more secure verification method for converged communication users with higher security levels can ensure the information security of converged communication users and improve the overall system security.

[0081] In some possible implementations, to increase login flexibility, a login assistance mechanism can be used for login control. For example, when a unified communication user A fails to perform security verification according to its corresponding target security verification method, the security management subsystem sends a friend-assisted login request to the unified communication user A. This request carries four friend accounts that are friends with the unified communication user A (the security management subsystem can pre-store friend relationships between unified communication users), wherein at least three of the four friend accounts have a higher security level than the unified communication user A. When the security management subsystem receives a friend-assisted response from the unified communication user A, this response carries an assistance login code corresponding to at least three of the four friend accounts. If the assistance login code corresponding to the at least three friend accounts is the same as the pre-stored corresponding assistance login code (indicating that the assistance login code carried in the friend-assisted response is a valid assistance login code), then the unified communication user A has obtained valid assistance from a friend, and the security management subsystem determines that the unified communication user A's security verification is successful.

[0082] Furthermore, for converged communication user A who logs in via the assisted login mechanism, the security management subsystem can forcibly log user A offline if the online time exceeds the security unit duration multiplied by the number of valid assisted login codes. For example, if the security unit duration is 30 minutes and the number of valid assisted login codes is 3, then if the online time of converged communication user A exceeds 90 minutes (30*3), the user will be forcibly logged off. Other cases follow the same logic.

[0083] In one possible example, the joint center subsystem is further configured to: determine the form of the corresponding feedback information based on the query method of each query; or, determine the form of the corresponding feedback information based on the voice content of each query; or, determine the form of the feedback information corresponding to each query based on the historical record of the preferred feedback information forms of each converged communication user; wherein the query method includes any one or more of sending information, voice call or IP call, and file transfer.

[0084] Specifically, to achieve the integration of multiple communication methods, the integration center subsystem can determine the form of the corresponding feedback information based on the query method of each query. For example, if the query is text, the feedback information can also be text; if the query is voice, the feedback information can also be voice, and so on. Additionally, the form of the feedback information can be determined based on the voice content of each query. For example, if the query is "Can you output my monthly detailed statement?", the feedback information could be a text statement. Another example is "Are there any nearby attractions recommended?", where the feedback information could be a link to a webpage introducing the attractions, pictures, videos, etc. Furthermore, the system can also query the historical records of the preferred feedback information formats for each integrated communication user to determine if that user has a preferred feedback format, such as voice or text, thereby determining the form of the feedback information corresponding to that user's query.

[0085] Among these, the voice or IP call in the question-and-answer mode is an automatic call made by the joint center subsystem, and it can communicate with the converged communication user based on the user's questions and answers. File transfer allows converged communication users to directly send a file to each other when they need to obtain it.

[0086] It is evident that adopting diverse standards to determine different forms of feedback information for different converged communication users can effectively improve the overall intelligence level of the customer service system and further realize converged communication.

[0087] In one possible example, the joint center subsystem is used to generate first feedback information corresponding to each inquiry if each inquiry has preset feedback information, including: if each inquiry has a preset feedback information, then using the preset feedback information as the first feedback information for each inquiry.

[0088] Specifically, when the joint center subsystem generates the first feedback information corresponding to the query information based on the preset feedback information, if there is only one preset feedback information for any query information, the preset feedback information can be directly used as the first feedback information for that query information.

[0089] It is evident that for queries with only one preset feedback message, directly using that preset feedback message as the first feedback message can effectively improve question-and-answer efficiency.

[0090] In one possible example, the joint center subsystem is configured to generate first feedback information corresponding to each query if each query has preset feedback information, including: if each query has at least two preset feedback messages, performing semantic analysis on the at least two preset feedback messages to obtain the core semantics of each preset feedback message; performing semantic analysis on each query to obtain the core semantics of each query; calculating the matching degree value between the core semantics of each query and the at least two preset feedback messages corresponding to each query, and using the preset feedback message corresponding to the highest matching degree value as the first feedback information for each query.

[0091] Specifically, taking any query as an example, if there are at least two preset feedback messages for that query, the joint center subsystem needs to select the preset feedback message with the highest matching degree from these two preset feedback messages as the first feedback message for that query. The matching degree value can be obtained as follows:

[0092] Semantic analysis is performed on at least two preset feedback messages to obtain the core semantics of each preset feedback message; semantic analysis is performed on the query message to obtain the core semantics of the query message; text similarity is calculated between the core semantics of the query message and the core semantics of each preset feedback message to obtain a similarity value, and the similarity value is used as the matching degree value.

[0093] It is evident that, for any given query, if at least two preset feedback messages exist, selecting the preset feedback message with the highest matching degree can effectively improve the accuracy of the feedback message.

[0094] In one possible example, the intelligent processing subsystem is configured to generate second feedback information based on each query, including: extracting common parameter information from each query and generating general feedback information based on the common parameter information; extracting feature parameter information from each query to obtain multiple feature parameter information; generating differentiated feedback information corresponding to each query based on each feature parameter information among the multiple feature parameter information; and combining the general feedback information and the differentiated feedback information to generate second feedback information.

[0095] Specifically, the intelligent processing subsystem may receive one or more query messages from the joint center subsystem. When there are multiple queries, to improve the efficiency of generating feedback information, common parameter information, such as identical or similar keywords, can be extracted from each query message. Then, general feedback information can be generated based on these common or similar keywords. Further, feature parameter information, such as differentiating keywords, can be extracted from each query message to obtain multiple feature parameter information. This feature parameter information can be used to generate differentiated feedback information, which is the most crucial component of the feedback information. Finally, the general feedback information and the differentiated feedback information are combined to generate the second feedback information.

[0096] In addition, for queries that cannot extract the same parameter information as other queries, as well as single queries, artificial intelligence (AI) technology can be used to analyze them separately and generate a second feedback message.

[0097] It is evident that the intelligent processing subsystem, by extracting parameters at different levels and then combining the parameters from each level for multiple queries, can effectively improve the efficiency of feedback information generation.

[0098] In one possible example, the intelligent processing subsystem is used to generate second feedback information by combining the general feedback information and the differentiated feedback information, including: inputting the general feedback information and the differentiated feedback information into the intelligent question-answering neural network, so that the intelligent question-answering neural network generates second feedback information based on the general feedback information and the differentiated feedback information.

[0099] Specifically, when the intelligent processing subsystem generates the second feedback information by combining general feedback information and differentiated feedback information, it can input the general feedback information and differentiated feedback information into the intelligent question-answering neural network, so that the intelligent question-answering neural network can generate the second feedback information based on the general feedback information and differentiated feedback information.

[0100] The intelligent question-answering neural network is obtained by pre-training a primary question-answering neural network until it converges. It can be any one or more of the following: Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), Deep Belief Neural Networks, Generative Adversarial Networks (GANs), Autoencoders (AEs), and Recurrent Neural Networks.

[0101] The convergence condition can be any one or more of the following: the loss value (i.e., the error) is less than a pre-set error threshold; or, the weight change (parameter) between two iterations is already very small, and a threshold can be set so that training stops when the weight change value is less than the parameter threshold; or, a maximum number of iterations can be set, and training stops when the number of iterations exceeds the maximum, which can be considered as reaching the model convergence condition. This improves the audio recognition capability and noise recognition capability of the initial denoising model.

[0102] Please see again Figure 4 This is an embodiment of the present invention of a fully automated intelligent customer service device based on converged communication. Figure 4 The diagram below illustrates the functional units of the intelligent customer service device 400 (hereinafter referred to as the intelligent customer service device). This embodiment of the application describes a fully automated intelligent customer service device 400 based on converged communication, which can be a built-in device or an external device of the aforementioned fully automated intelligent customer service system based on converged communication. The security management subsystem of the aforementioned system can correspond to the security management unit of this device; the joint center subsystem can correspond to the joint center unit of this device; and the intelligent processing subsystem can correspond to the intelligent processing unit of this device.

[0103] In one implementation of the apparatus according to an embodiment of the present invention, the apparatus includes:

[0104] Security management unit 410 is used to verify whether each of the multiple converged communication users meets the login conditions, and if each converged communication user meets the login conditions, allow each converged communication user to log in to the joint center unit;

[0105] The joint central unit 420 includes multiple node devices, used to determine the target node device among the multiple node devices for intelligent question answering.

[0106] The joint central unit 420 is also configured to instruct the target node device to receive each query message sent by each converged communication user and generate a first feedback message corresponding to each query message;

[0107] The joint central unit 420 is also configured to send each query information to the intelligent processing unit when the target node device has not generated the first feedback information;

[0108] The intelligent processing subunit 430 is used to generate second feedback information based on each query information and send the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user.

[0109] In one possible example, the security management unit 410, in verifying whether each of the multiple converged communication users meets the login conditions, specifically includes: obtaining the security level of each of the multiple converged communication users, and determining a target security verification method based on the security level of each converged communication user, the target security verification method including a first verification method and a second verification method; and verifying whether each converged communication user meets the login conditions based on the target security verification method.

[0110] In one possible example, the security management unit 410, in verifying whether each converged communication user meets the login conditions according to the target security verification method, specifically includes: if the target security verification method of the converged communication user is the first verification method, then verifying the login account information and login password of the converged communication user; and if the login account information and login password of the converged communication user match, determining that the converged communication user has passed the verification.

[0111] In one possible example, the security management unit 410, in verifying whether each converged communication user meets the login conditions according to the target security verification method, specifically includes: if the target security verification method of the converged communication user is the second verification method, then verifying the converged communication user's login account information and login password; and if the converged communication user's login account information and login password match, obtaining a pre-stored public key based on the converged communication user's login account information or login password, generating a standard private key based on the public key according to a specific encryption method, and sending the public key to the converged communication user so that the converged communication user generates a private key to be verified based on the public key; receiving the private key to be verified sent by the converged communication user, comparing whether the private key to be verified matches the standard private key, and determining that the converged communication user has passed verification if the private key to be verified matches the standard private key.

[0112] In one possible example, the joint central unit 420 is further configured to: determine the form of the corresponding feedback information based on the query method of each query information; or, determine the form of the corresponding feedback information based on the voice content of each query information; or, determine the form of the feedback information corresponding to each query information based on the historical record of the preferred feedback information forms of each converged communication user; wherein the query method includes any one or more of sending information, voice call or IP call, and file transfer.

[0113] In one possible example, the joint central unit 420 is further configured to instruct the target node device to receive each query message sent by each converged communication user and generate first feedback information corresponding to each query message, including: extracting at least one keyword from each query message to obtain at least one keyword for each query message; querying whether there is preset feedback information for the corresponding query message based on the at least one keyword; if there is preset feedback information for each query message, generating first feedback information corresponding to the query message based on the preset feedback information.

[0114] In one possible example, the joint central unit 420, in generating first feedback information corresponding to each inquiry information based on the preset feedback information if each inquiry information has preset feedback information, specifically includes: if each inquiry information has a preset feedback information, then using the preset feedback information as the first feedback information for each inquiry information.

[0115] In one possible example, the joint central unit 420, in generating first feedback information corresponding to each query information based on the preset feedback information if each query information has preset feedback information, specifically includes: if each query information has at least two preset feedback information, performing semantic analysis on the at least two preset feedback information to obtain the core semantics of each preset feedback information; performing semantic analysis on each query information to obtain the core semantics of each query information; calculating the matching degree value between the core semantics of each query information and the at least two preset feedback information corresponding to each query information, and taking the preset feedback information corresponding to the highest matching degree value as the first feedback information of each query information.

[0116] In one possible example, the intelligent processing unit 430, in generating second feedback information based on each query information, specifically includes: extracting common parameter information from each query information respectively, and generating general feedback information based on the common parameter information; extracting feature parameter information from each query information respectively to obtain multiple feature parameter information; generating differentiated feedback information corresponding to each query information based on each feature parameter information among the multiple feature parameter information; and combining the general feedback information and the differentiated feedback information to generate second feedback information.

[0117] In one possible example, the intelligent processing unit 430, in generating second feedback information by combining the general feedback information and the differentiated feedback information, specifically includes: inputting the general feedback information and the differentiated feedback information into the intelligent question-answering neural network, so that the intelligent question-answering neural network generates second feedback information based on the general feedback information and the differentiated feedback information.

[0118] In some embodiments, the fully automated intelligent customer service device based on converged communication may further include input / output interfaces, communication interfaces, power supplies, and communication buses.

[0119] This application embodiment can divide the fully automated intelligent customer service device based on converged communication into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0120] Please see again Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. The electronic device includes a power supply module and other structures, and includes a processor 501, a storage device 502, and a communication interface 503. The processor 501, the storage device 502, and the communication interface 503 can exchange data.

[0121] The storage device 502 may include volatile memory, such as random-access memory (RAM); the storage device 502 may also include non-volatile memory, such as flash memory, solid-state drive (SSD), etc.; the storage device 502 may also include combinations of the above types of memory. The communication interface 503 is an interface for data exchange between internal devices of the wireless headset, such as between the storage device 502 and the processor 501.

[0122] The processor 501 may be a central processing unit (CPU). In one embodiment, the processor 501 may also be a graphics processing unit (GPU). The processor 501 may also be a combination of a CPU and a GPU. In one embodiment, the storage device 502 is used to store program instructions. The processor 501 can invoke the program instructions to perform the following steps:

[0123] Verify whether each of the multiple converged communication users meets the login requirements, and if each converged communication user meets the login requirements, allow each converged communication user to log in to the joint center subsystem;

[0124] The target node device among the plurality of node devices is identified for intelligent question answering;

[0125] The target node device is instructed to receive each query message sent by each converged communication user and generate a first feedback message corresponding to each query message;

[0126] When the target node device does not generate the first feedback information, each query information is sent to the intelligent processing subsystem;

[0127] A second feedback message is generated based on each query message, and the second feedback message is sent to the joint center subsystem so that the joint center subsystem forwards the second feedback message to the converged communication user.

[0128] In one possible example, the processor 501, in verifying whether each of the multiple converged communication users meets the login conditions, specifically includes: obtaining the security level of each of the multiple converged communication users, and determining a target security verification method based on the security level of each converged communication user, the target security verification method including a first verification method and a second verification method; and verifying whether each converged communication user meets the login conditions based on the target security verification method.

[0129] In one possible example, the processor 501, in verifying whether each converged communication user meets the login conditions according to the target security verification method, specifically includes: if the target security verification method of the converged communication user is the first verification method, then verifying the login account information and login password of the converged communication user; and if the login account information and login password of the converged communication user match, determining that the converged communication user has passed the verification.

[0130] In one possible example, the processor 501, in verifying whether each converged communication user meets the login conditions according to the target security verification method, specifically includes: if the target security verification method of the converged communication user is the second verification method, then verifying the converged communication user's login account information and login password; and if the converged communication user's login account information and login password match, obtaining a pre-stored public key based on the converged communication user's login account information or login password, generating a standard private key based on the public key using a specific encryption method, and sending the public key to the converged communication user so that the converged communication user generates a private key to be verified based on the public key; receiving the private key to be verified sent by the converged communication user, comparing whether the private key to be verified matches the standard private key, and determining that the converged communication user has passed verification if the private key to be verified matches the standard private key.

[0131] In one possible example, the processor 501 is further configured to: determine the form of the corresponding feedback information based on the query method of each query information; or, determine the form of the corresponding feedback information based on the voice content of each query information; or, determine the form of the feedback information corresponding to each query information based on the historical record of the preferred feedback information forms of each converged communication user; wherein the query-response method includes any one or more of sending information, voice call or IP call, and file transfer.

[0132] In one possible example, the processor 501 is further configured to instruct the target node device to receive each query message sent by each converged communication user and generate first feedback information corresponding to each query message, including: extracting at least one keyword from each query message to obtain at least one keyword for each query message; querying whether preset feedback information exists for the corresponding query message based on the at least one keyword; and generating first feedback information corresponding to the query message based on the preset feedback information if preset feedback information exists for each query message.

[0133] In one possible example, the processor 501, in generating first feedback information corresponding to the query information based on the preset feedback information if each query information has preset feedback information, specifically includes: if each query information has a preset feedback information, then using the preset feedback information as the first feedback information for each query information.

[0134] In one possible example, the processor 501, in generating first feedback information corresponding to each query information based on the preset feedback information if each query information has preset feedback information, specifically includes: if each query information has at least two preset feedback information, performing semantic analysis on the at least two preset feedback information to obtain the core semantics of each preset feedback information; performing semantic analysis on each query information to obtain the core semantics of each query information; calculating the matching degree value between the core semantics of each query information and the at least two preset feedback information corresponding to each query information, and using the preset feedback information corresponding to the highest matching degree value as the first feedback information of each query information.

[0135] In one possible example, the processor 501, in generating second feedback information based on each query information, specifically includes: extracting common parameter information from each query information respectively, and generating general feedback information based on the common parameter information; extracting feature parameter information from each query information respectively to obtain multiple feature parameter information; generating differentiated feedback information corresponding to each query information based on each feature parameter information among the multiple feature parameter information; and combining the general feedback information and the differentiated feedback information to generate second feedback information.

[0136] In one possible example, the processor 501, in generating second feedback information by combining the general feedback information and the differentiated feedback information, specifically includes: inputting the general feedback information and the differentiated feedback information into the intelligent question-answering neural network, so that the intelligent question-answering neural network generates second feedback information based on the general feedback information and the differentiated feedback information.

[0137] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments.

[0138] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform some or all of the steps of any of the methods described in the above method embodiments.

[0139] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0140] The above-disclosed embodiments are merely some examples of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for fully automated intelligent customer service processing, characterized in that, The method includes: The security management subsystem verifies whether each of the multiple converged communication users meets the login requirements, and if each converged communication user meets the login requirements, allows each converged communication user to log in to the joint center subsystem; The joint center subsystem identifies the target node device among the multiple node devices included for intelligent question answering; the node devices are in a distributed connection state, capable of jointly storing important data, but capable of separately executing customer service inquiry operations. The joint center subsystem instructs the target node device to receive each query message sent by each converged communication user and generate a first feedback message corresponding to each query message; When the target node device fails to generate the first feedback information, the joint center subsystem sends each query information to the intelligent processing subsystem. The intelligent processing subsystem generates second feedback information based on each query information and sends the second feedback information to the joint center subsystem, so that the joint center subsystem forwards the second feedback information to the converged communication user; The security management subsystem is used to verify whether each of the multiple converged communication users meets the login conditions, including: obtaining the security level of each of the multiple converged communication users, and determining a target security verification method based on the security level of each converged communication user, wherein the target security verification method includes a first verification method and a second verification method; and verifying whether each converged communication user meets the login conditions based on the target security verification method. The security management subsystem is used to verify whether each converged communication user meets the login conditions according to the target security verification method, including: If the target security verification method for the converged communication user is the second verification method, then the login account information and login password of the converged communication user are verified; and if the login account information and login password of the converged communication user match, a pre-stored public key is obtained according to the login account information or login password of the converged communication user, a standard private key is generated according to the public key using a specific encryption method, and the public key is sent to the converged communication user so that the converged communication user can generate a private key to be verified according to the public key; the private key to be verified sent by the converged communication user is received, and the private key to be verified is compared with the standard private key. If the private key to be verified matches the standard private key, it is determined that the converged communication user has passed the verification. Specifically, when user A fails to perform security verification according to its corresponding target security verification method, the security management subsystem sends a friend assistance login request to user A. The friend assistance request carries four friend accounts that are friends with user A, and at least three of the four friend accounts have a higher security level than user A. When the security management subsystem receives a friend assistance response from user A, the friend assistance response carries an assistance login code corresponding to at least three of the four friend accounts. If the assistance login code corresponding to the at least three friend accounts is the same as the pre-stored corresponding assistance login code, it is determined that user A has obtained valid assistance from friends, and the security management subsystem determines that user A's security verification is successful. For user A who logs in through the assistance login mechanism, if user A's online time exceeds the security unit time multiplied by the valid assistance login code, the security management subsystem will forcibly log user A offline.

2. The method according to claim 1, characterized in that, The security management subsystem verifies whether each converged communication user meets the login requirements according to the target security verification method, including: If the target security verification method for the converged communication user is the first verification method, then the login account information and login password of the converged communication user are verified; and if the login account information and login password of the converged communication user match, it is determined that the converged communication user has passed the verification.

3. The method according to claim 1, characterized in that, The method further includes: the joint center subsystem determining the form of the corresponding feedback information based on the query method of each query information; or, determining the form of the corresponding feedback information based on the voice content of each query information; or, determining the form of the feedback information corresponding to each query information based on the historical record of the preferred feedback information forms of each converged communication user; wherein, the query method includes any one or more of sending information, voice call or IP call, and file transfer.

4. The method according to claim 1, characterized in that, The method includes: the joint center subsystem instructing the target node device to receive each query message sent by each converged communication user, and generating first feedback information corresponding to each query message, including: extracting at least one keyword from each query message to obtain at least one keyword for each query message; querying whether there is preset feedback information for the corresponding query message based on the at least one keyword; if there is preset feedback information for each query message, generating first feedback information corresponding to the query message based on the preset feedback information.

5. The method according to claim 4, characterized in that, If each query information has preset feedback information, the joint center subsystem generates first feedback information corresponding to the query information based on the preset feedback information, including: if each query information has a preset feedback information, the preset feedback information is used as the first feedback information for each query information.

6. The method according to claim 5, characterized in that, The joint center subsystem is used to generate first feedback information corresponding to each inquiry if each inquiry has preset feedback information. This includes: if each inquiry has at least two preset feedback messages, performing semantic analysis on the at least two preset feedback messages to obtain the core semantics of each preset feedback message; performing semantic analysis on each inquiry to obtain the core semantics of each inquiry; calculating the matching degree value between the core semantics of each inquiry and the at least two preset feedback messages corresponding to each inquiry, and using the preset feedback information corresponding to the highest matching degree value as the first feedback information for each inquiry.

7. A computer-readable storage medium, characterized in that, Used to store computer program instructions for implementing the method according to any one of claims 1 to 6.

8. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium. When the computer instructions are executed by the processor of a computer device, they perform the method described in any one of claims 1 to 6.

Images