Intelligent interaction method and system for distributed multi-mode online customer service

Abstract

The invention relates to the technical field of retrieval feedback calculation, in particular to a distributed multi-mode online customer service intelligent interaction method and system. The scheme comprises the following steps: setting an infrastructure layer, and completing the construction of infrastructures; after the infrastructure layer obtains the data, online knowledge base management is carried out; setting course information management on the basis of online knowledge base management; performing time statistics on all historical question and answer information; according to a time statistical result, performing quick answer retrieval according to a question proposed by a client user; and detailed billboard display is carried out. According to the scheme, the dependence on manual customer service is reduced through automatic processing, and the labor cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of retrieval feedback computing technology, and more specifically, to an intelligent interaction method and system for distributed multimodal online customer service. Background Art

[0002] The retrieval and question-answering functions of the customer service system are the bridges for modern enterprises to communicate with their customers. Their importance is not only reflected in improving customer satisfaction, but also in helping enterprises reduce costs and improve efficiency. The retrieval and question-answering system can quickly find answers to questions from a large amount of data and provide customers with instant feedback. The automated question-answering system reduces customer waiting time and improves customer satisfaction. By automatically handling common problems, the reliance on manual customer service is reduced, thereby reducing labor costs.

[0003] Prior to the present invention, the existing traditional customer service system was mainly based on retrieval technology, with limited ability to understand the semantics of questions raised by users and generate responses. It could only answer questions according to fixed rules, and its interaction mode was single, making it difficult to meet the complex and diverse questioning needs of users. In addition, the construction of its knowledge base mainly relied on manually written question and answer pairs, which consumed a lot of time and energy, making the construction of the knowledge base cumbersome. Summary of the invention

[0004] In view of the above problems, the present invention proposes an intelligent interaction method and system for distributed multimodal online customer service, which reduces the dependence on manual customer service and reduces labor costs through automated processing.

[0005] According to a first aspect of an embodiment of the present invention, a distributed multimodal online customer service intelligent interaction method is provided.

[0006] In one or more embodiments, preferably, the intelligent interaction method of distributed multimodal online customer service includes:

[0007] Set up the infrastructure layer to complete the construction of the infrastructure;

[0008] After acquiring data at the infrastructure layer, online knowledge base management is performed;

[0009] Based on the online knowledge base management, set up course information management;

[0010] Perform time statistics on all historical question and answer information;

[0011] According to the time statistics, quickly search for answers to questions raised by client users;

[0012] Make a detailed kanban presentation.

[0013] In one or more embodiments, preferably, the setting of the infrastructure layer to complete the construction of the infrastructure specifically includes:

[0014] Set up large model servers and application servers;

[0015] Set up the K8S containerized platform;

[0016] Set up the MySQL database;

[0017] Set up a low-code platform.

[0018] In one or more embodiments, preferably, after acquiring data at the infrastructure layer, online knowledge base management is performed, specifically including:

[0019] Setting up knowledge base management includes creating, viewing, searching, logging off and clearing functions;

[0020] Under the New Function, click the Create QA button, a pop-up window will pop up, enter the question and answer fields, and click OK to save;

[0021] Under the View function, click the View Details button to pop up a pop-up window that displays field information such as the question ID and data source, which cannot be edited;

[0022] Under the search function, enter the conditions in the problem description and date box, click the search button, the system will search according to the conditions, and the page will display the search results;

[0023] Under the Clear function, click the Clear button to clear the search conditions in the condition input box; under the Offline function, click the Offline button and a pop-up window will appear, asking if you are sure to go offline. Click OK and the data will enter the pending review interface.

[0024] In one or more embodiments, preferably, based on the online knowledge base management, course information management is set, which specifically includes:

[0025] The page includes new creation, deletion, editing, and search functions;

[0026] To create a new function, you need to fill in the course id and course name;

[0027] Click the Edit button in the Operation column to enter the editing interface and modify the original course information;

[0028] Click the Delete button to delete the data;

[0029] The page supports fuzzy search based on course name.

[0030] In one or more embodiments, preferably, the time statistics of all historical question and answer information specifically include:

[0031] Set the statistical categories to monthly statistics, quarterly statistics and yearly statistics;

[0032] Statistics are collected based on the number of fuzzy searches and exports of the followed users;

[0033] Each statistic includes the total number of questions and answers, effectiveness, like rate, and failure rate fields.

[0034] In one or more embodiments, preferably, the method of performing a quick answer search according to the time statistics result and the question raised by the client user specifically includes:

[0035] The vocabulary management page includes functions such as search by sensitive word type, fuzzy search by sensitive word, create, edit, and delete;

[0036] Search by sensitive word type to form the first number;

[0037] Fuzzy search based on sensitive words to generate a second number;

[0038] Get the questions raised by the client user, perform vocabulary extraction, and determine whether there are sensitive words;

[0039] If there is a sensitive word, the first calculation formula is used to calculate the current most matching sensitive word feedback number;

[0040] Feedback the preset feedback answer based on the current most matching sensitive word feedback number;

[0041] Get all current matching words, get the time group where the current word is located, the time minute is preset, and the time group is divided into one of monthly statistics, quarterly statistics and yearly statistics;

[0042] Performing validity calculation using a second calculation formula according to the fuzzy search and the number of derivation times;

[0043] Based on all current matching words, use the big model to obtain the corresponding optimal feedback list under the current knowledge base and course;

[0044] When the effectiveness does not exceed the preset value, a comprehensive index is calculated using a third calculation formula according to the optimal feedback list and the total number of answers, effectiveness rate, like rate and failure rate fields corresponding to the answers in the optimal feedback list;

[0045] When the effectiveness exceeds a preset value, the associated feedback of the answer associated with the current search in the exported data is incorporated into the optimal feedback list;

[0046] And calculate the comprehensive index using the third calculation formula based on the total number of answers, efficiency, like rate and failure rate fields;

[0047] The result of the comprehensive index is used as the first feedback result. When the effectiveness exceeds a preset value, the answer with the highest comprehensive index in the exported data associated with the current search is used as the second feedback result.

[0048] The first calculation formula is:

[0049] B=k1×S1+k2×S2

[0050] Wherein, B is the current most matched sensitive word feedback number, k1 is the first preset coefficient, k2 is the second preset coefficient, S1 is the first number, and S2 is the second number;

[0051] The second calculation formula is:

[0052] Y=D / M

[0053] Among them, Y is validity, D is the number of exports, and M is fuzzy search;

[0054] The third calculation formula is:

[0055] ZZ=DZ×xl×(1-sb) 3 ×(1+di anz)

[0056] Among them, ZZ is the comprehensive index, DZ is the total number of answers, xl is the effective rate, dianz is the like rate, and sb is the failure rate.

[0057] In one or more embodiments, preferably, the performing of detailed billboard display specifically includes:

[0058] Set up four follow modules, namely: follow user selection, data statistics, line chart and data report;

[0059] You can choose to view all data or the data of a certain followed user as the selected user;

[0060] Set data statistics including total number of questions and answers, total number of questions, number of visits, total number of likes and dislikes, efficiency, like rate and failure rate;

[0061] Count the number of questions and answers and the number of questions every day in the past 5 days and make a line graph;

[0062] Count the number and proportion of all likes and dislikes as a data report.

[0063] According to a second aspect of an embodiment of the present invention, a distributed multimodal online customer service intelligent interaction system is provided.

[0064] In one or more embodiments, preferably, the distributed multimodal online customer service intelligent interaction system includes:

[0065] The infrastructure setting module is used to set up the infrastructure layer and complete the construction of the infrastructure;

[0066] The knowledge base management module is used to perform online knowledge base management after acquiring data at the infrastructure layer;

[0067] The course management module is used to set up course information management based on the online knowledge base management;

[0068] Time statistics module, used to perform time statistics on all historical question and answer information;

[0069] Intelligent recommendation module, used to quickly retrieve answers to questions raised by client users based on time statistics results;

[0070] The kanban display module is used for detailed kanban display.

[0071] According to a third aspect of an embodiment of the present invention, there is provided a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method as described in any one of the first aspect of the embodiment of the present invention is implemented.

[0072] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement any one of the methods described in the first aspect of the embodiment of the present invention.

[0073] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0074] In the solution of the present invention, an automated detection question-answering interactive system is provided to reduce labor costs.

[0075] In the solution of the present invention, according to the time statistics results, a quick answer search is performed according to the questions raised by the client user to achieve a quick question-answer interaction.

[0076] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0077] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0079] Figure 1 It is a flow chart of an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0080] Figure 2 The present invention is a flowchart of setting up an infrastructure layer in a distributed multimodal online customer service intelligent interaction method and completing the construction of an infrastructure.

[0081] Figure 3 It is a flowchart of performing online knowledge base management after acquiring data at the infrastructure layer in an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0082] Figure 4 The present invention is a flowchart of setting course information management based on online knowledge base management in a distributed multimodal online customer service intelligent interaction method according to an embodiment of the present invention.

[0083] Figure 5 It is a flowchart of performing time statistics on all historical question and answer information in an intelligent interaction method of a distributed multimodal online customer service according to an embodiment of the present invention.

[0084] Figure 6 The present invention is a flowchart of a distributed multimodal online customer service intelligent interaction method for quickly retrieving answers to questions raised by client users based on time statistics results in one embodiment of the present invention.

[0085] Figure 7 It is a flowchart of performing detailed kanban display in an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0086] Figure 8 It is a structural diagram of an intelligent interactive system of distributed multimodal online customer service according to an embodiment of the present invention.

[0087] Fig. 9 It is a structural diagram of an electronic device in one embodiment of the present invention. DETAILED DESCRIPTION

[0088] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0089] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0090] The retrieval and question-answering functions of the customer service system are the bridges for modern enterprises to communicate with their customers. Their importance is not only reflected in improving customer satisfaction, but also in helping enterprises reduce costs and improve efficiency. The retrieval and question-answering system can quickly find answers to questions from a large amount of data and provide customers with instant feedback. The automated question-answering system reduces customer waiting time and improves customer satisfaction. By automatically handling common problems, the reliance on manual customer service is reduced, thereby reducing labor costs.

[0091] Prior to the present invention, the existing traditional customer service system was mainly based on retrieval technology, with limited ability to understand the semantics of questions raised by users and generate responses. It could only answer questions according to fixed rules, and its interaction mode was single, making it difficult to meet the complex and diverse questioning needs of users. In addition, the construction of its knowledge base mainly relied on manually written question and answer pairs, which consumed a lot of time and energy, making the construction of the knowledge base cumbersome.

[0092] In an embodiment of the present invention, a distributed multi-modal online customer service intelligent interaction method and system are provided. The solution reduces the reliance on manual customer service through automated processing and reduces labor costs.

[0093] According to a first aspect of an embodiment of the present invention, a distributed multimodal online customer service intelligent interaction method is provided.

[0094] Figure 1 It is a flow chart of an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0095] In one or more embodiments, preferably, the intelligent interaction method of distributed multimodal online customer service includes:

[0096] S101, set up the infrastructure layer and complete the construction of the infrastructure;

[0097] S102, after acquiring data at the infrastructure layer, perform online knowledge base management;

[0098] S103. Set up course information management based on online knowledge base management;

[0099] S104, performing time statistics on all historical question and answer information;

[0100] S105, based on the time statistics result, quickly search for answers according to the questions raised by the client user;

[0101] S106. Perform detailed billboard display.

[0102] In the embodiment of the present invention, by automatically handling common problems, the reliance on manual customer service is reduced, thereby reducing labor costs; in addition, the automated system can handle inquiries from multiple customers at the same time, thereby improving the efficiency of handling problems. In addition, it should be noted that multimodal online customer service can import ASR text (subtitle information).

[0103] Figure 2 The present invention is a flowchart of setting up an infrastructure layer in a distributed multimodal online customer service intelligent interaction method and completing the construction of an infrastructure.

[0104] like Figure 2 As shown, in one or more embodiments, preferably, the setting of the infrastructure layer to complete the construction of the infrastructure specifically includes:

[0105] S201, setting up a large model server and an application server;

[0106] S202. Set up the K8S container platform;

[0107] S203, setting up MySQL database;

[0108] S204. Set up a low-code platform.

[0109] In the embodiment of the present invention, the infrastructure layer is the foundation of the system and provides the operating environment and support for other layers. It includes hardware, operating system, network infrastructure, storage system, etc. At this layer, the focus is mainly on the reliability, scalability, security and performance of the system. The design decisions of the infrastructure layer will directly affect the overall stability and availability of the system. In this architecture, we used basic technologies such as large model server, application server, operating system, large model middleware service, low code platform, MySQL database, k8s, docker, etc.

[0110] Figure 3 It is a flowchart of performing online knowledge base management after acquiring data at the infrastructure layer in an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0111] like Figure 3 As shown, in one or more embodiments, preferably, after acquiring data at the infrastructure layer, online knowledge base management is performed, specifically including:

[0112] S301. Set up knowledge base management including creating, viewing, searching, logging off and clearing functions;

[0113] S302. Under the new function, click the Create QA button, a pop-up window will pop up, enter the question and answer fields, and click OK to save;

[0114] S303. Under the View function, click the View Details button, and a pop-up window will pop up, which displays the question ID, data source and other field information, which cannot be edited;

[0115] S304. Under the search function, enter conditions in the problem description and date box, click the search button, the system will search according to the conditions, and the page will display the search results;

[0116] S305. Under the Clear function, click the Clear button to clear the search conditions in the condition input box; under the Offline function, click the Offline button, a pop-up window will appear, asking if you are sure to go offline, click OK, and the data will enter the pending review interface.

[0117] In an embodiment of the present invention, the knowledge base required for answering can be uniformly managed in the background, and the knowledge base is divided into FAQ library and knowledge base. FAQ library management-published page, the page mainly includes the functions of creating, viewing, searching, offline, and clearing; click the Create QA button, a pop-up window pops up, enter the question and answer fields, and click OK to save; click the View Details button, a pop-up window pops up, the pop-up window displays the question ID, data source and other field information, which cannot be edited; enter the conditions in the question description and date box, click the Search button, the system will query according to the conditions, and the page will display the query results; click the Clear button, the search conditions in the condition input box will be cleared; click the Offline button, a pop-up window appears, whether to confirm the offline, click OK, the data enters the pending review interface; Knowledge Base-FAQ Library Management-Pending Review Page, the page includes the functions of viewing, publishing, and deleting; click the View button to enter the View page, the page displays the question ID, question description, reviewer and other fields, and supports the modification function. Click the Publish button, a pop-up window appears, the pop-up window content shows whether to publish, click OK to publish to the published page, the Publish button in the View page has the same function as this button. Click the Delete button and a pop-up window will appear. The pop-up window content shows whether to delete. Click OK to delete the data.

[0118] Figure 4 The present invention is a flowchart of setting course information management based on online knowledge base management in a distributed multimodal online customer service intelligent interaction method according to an embodiment of the present invention.

[0119] like Figure 4 As shown, in one or more embodiments, preferably, based on the online knowledge base management, course information management is set, specifically including:

[0120] S401, the page includes the functions of creating, deleting, editing and searching;

[0121] S402, the new function needs to fill in the course id and course name;

[0122] S403, click the edit button in the operation column to enter the editing interface and modify the original course information;

[0123] S404, click the delete button to delete the data;

[0124] S405. The page supports fuzzy search based on course name.

[0125] In the embodiment of the present invention, the backend management system - course information management page contains the functions of creating, deleting, editing, and searching. The page supports fuzzy search based on the course name; the new function requires filling in the course ID and course name, and selecting the user to be followed (the data source is the management of the user to be followed). Click the edit button in the operation column to enter the editing interface, where you can modify the original course information; click the delete button to delete the data.

[0126] Figure 5 It is a flowchart of performing time statistics on all historical question and answer information in an intelligent interaction method of a distributed multimodal online customer service according to an embodiment of the present invention.

[0127] like Figure 5 As shown, in one or more embodiments, preferably, the time statistics of all historical question and answer information specifically include:

[0128] S501, setting the statistical categories to monthly statistics, quarterly statistics and yearly statistics;

[0129] S502, collecting statistics based on the number of fuzzy searches and exports of the followed users;

[0130] S503. Each statistic includes the fields of total number of questions and answers, efficiency, like rate, and failure rate.

[0131] In an embodiment of the present invention, statistics include monthly statistics, quarterly statistics, and yearly statistics. Fuzzy search and export can be performed on the statistics based on the followed users. Each statistic includes the fields of total number of questions and answers, efficiency, like rate, and failure rate. The statistical time field is counted by month, quarter, and year respectively.

[0132] Figure 6 The present invention is a flowchart of a distributed multimodal online customer service intelligent interaction method for quickly retrieving answers to questions raised by client users based on time statistics results in one embodiment of the present invention.

[0133] like Figure 6 As shown, in one or more embodiments, preferably, the method of performing a quick answer search according to the time statistics result and the question raised by the client user specifically includes:

[0134] S601, setting the vocabulary management page to include functions of query by sensitive word type, fuzzy query by sensitive word, create, edit, and delete;

[0135] S602, searching by sensitive word type to form a first number;

[0136] S603, fuzzy search based on sensitive words to generate a second number;

[0137] S604: Obtain the question raised by the client user, perform vocabulary extraction, and determine whether there are sensitive words;

[0138] S605: If there is a sensitive word, use the first calculation formula to calculate the current most matching sensitive word feedback number;

[0139] S606: Feedback a preset feedback answer according to the current most matching sensitive word feedback number;

[0140] S607, obtaining all current matching words, obtaining the time grouping where the current word is located, the time minute is preset, and the time grouping is divided into one of monthly statistics, quarterly statistics and yearly statistics;

[0141] S608, performing validity calculation using a second calculation formula according to the fuzzy search and the number of derivation times;

[0142] S609, using the big model to obtain the corresponding optimal feedback list under the current knowledge base and course according to all current matching words;

[0143] S6010, when the effectiveness does not exceed the preset value, a comprehensive index is calculated using a third calculation formula according to the optimal feedback list and the total number of answers, effectiveness rate, like rate and failure rate fields corresponding to the answers in the optimal feedback list;

[0144] S6011, when the validity exceeds a preset value, the associated feedback of the answer associated with the current search in the exported data is incorporated into the optimal feedback list;

[0145] S6012, and calculating the comprehensive index using the third calculation formula according to the total number of answers, efficiency, like rate and failure rate fields;

[0146] S6013, taking the result of the comprehensive index as the first feedback result, and when the effectiveness exceeds a preset value, taking the answer with the highest comprehensive index in the exported data associated with the current search as the second feedback result;

[0147] The first calculation formula is:

[0148] B=k1×S1+k2×S2

[0149] Wherein, B is the current most matched sensitive word feedback number, k1 is the first preset coefficient, k2 is the second preset coefficient, S1 is the first number, and S2 is the second number;

[0150] The second calculation formula is:

[0151] Y=D / M

[0152] Among them, Y is validity, D is the number of exports, and M is fuzzy search;

[0153] The third calculation formula is:

[0154] ZZ=DZ×xl×(1-sb) 3 ×(1+di anz)

[0155] Among them, ZZ is the comprehensive index, DZ is the total number of answers, xl is the effective rate, dianz is the like rate, and sb is the failure rate.

[0156] In an embodiment of the present invention, the vocabulary management page includes functions such as query by sensitive word type, fuzzy query by sensitive word, create, edit, and delete; query by sensitive word type; fuzzy query by sensitive word; create: pull down to select sensitive word type, enter sensitive word and status Edit: click the edit button in the operation column to enter the editing interface, where you can modify the original vocabulary information; delete: click the delete button to delete the data. For questions and answers missing in the knowledge base, you can temporarily supplement them in business management-background management-automatic reply management. When asking questions and answering, you will first filter whether the questions are questions in the automatic reply management. Query logic: Get questions raised by client users; search the corpus in the automatic reply management based on the questions; get the standard answers to the questions based on the corpus.

[0157] Figure 7 It is a flowchart of performing detailed kanban display in an intelligent interaction method of distributed multimodal online customer service according to an embodiment of the present invention.

[0158] like Figure 7 As shown, in one or more embodiments, preferably, the detailed billboard display specifically includes:

[0159] S701, setting four attention modules, namely: followed user selection, data statistics, line chart and data report;

[0160] S702, providing a user who is being followed with the option to view all data or data under a certain user as a choice of the user being followed;

[0161] S703, setting data statistics including total number of questions and answers, total number of questions, number of visits, total number of likes and dislikes, efficiency, like rate and failure rate;

[0162] S704, counting the number of questions and answers and the number of questions in the past five days and making a line graph;

[0163] S705. Count the number and proportion of all likes and dislikes as a data report.

[0164] In an embodiment of the present invention, in order to facilitate managers to view system data, a dashboard function is provided. This page is divided into four modules, namely, followed user selection, data statistics, line chart, and data report; followed user selection: you can choose to view all data or data under a certain followed user; data statistics: including the total number of questions and answers, the total number of questions, the number of visits, the total number of likes and dislikes, the efficiency, the like rate, and the failure rate; line chart: statistics the number of questions and answers and the number of questions every day in the past 5 days; data report: statistics the number and proportion of all likes and dislikes.

[0165] According to a second aspect of an embodiment of the present invention, a distributed multimodal online customer service intelligent interaction system is provided.

[0166] Figure 8 It is a structural diagram of an intelligent interactive system of distributed multimodal online customer service according to an embodiment of the present invention.

[0167] In one or more embodiments, preferably, the distributed multimodal online customer service intelligent interaction system includes:

[0168] The infrastructure setting module 801 is used to set up the infrastructure layer and complete the construction of the infrastructure;

[0169] The knowledge base management module 802 is used to perform online knowledge base management after acquiring data at the infrastructure layer;

[0170] The course management module 803 is used to set up course information management based on the online knowledge base management;

[0171] The time statistics module 804 is used to perform time statistics on all historical question and answer information;

[0172] Intelligent recommendation module 805, used to quickly retrieve answers according to questions raised by client users based on time statistics results;

[0173] The kanban display module 806 is used for performing detailed kanban display.

[0174] In the embodiment of the present invention, a system suitable for different structures is realized through a series of modular designs. The system can achieve closed-loop, reliable and efficient execution through collection, analysis and control.

[0175] According to a third aspect of an embodiment of the present invention, there is provided a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method as described in any one of the first aspect of the embodiment of the present invention is implemented.

[0176] According to a fourth aspect of an embodiment of the present invention, an electronic device is provided. Fig. 9It is a structural diagram of an electronic device in one embodiment of the present invention. Fig. 9 The electronic device shown is an intelligent interactive device for universal distributed multimodal online customer service. The electronic device can be a smart phone, tablet computer, or other device. As shown, the electronic device 900 includes a processor 901 and a memory 902. The processor 901 is electrically connected to the memory 902. The processor 901 is the control center of the terminal 900, which uses various interfaces and lines to connect various parts of the entire terminal, and executes various functions of the terminal and processes data by running or calling computer programs stored in the memory 902, and calling data stored in the memory 902, thereby monitoring the terminal as a whole.

[0177] In this embodiment, the processor 901 in the electronic device 900 will load the instructions corresponding to the processes of one or more computer programs into the memory 902 according to the following steps, and the processor 901 will run the computer programs stored in the memory 902 to achieve various functions: setting up an infrastructure layer to complete the construction of the infrastructure; after obtaining data at the infrastructure layer, performing online knowledge base management; setting up course information management based on the online knowledge base management; performing time statistics on all historical question and answer information; based on the time statistics results, performing rapid answer retrieval according to the questions raised by the client users; and performing detailed dashboard display.

[0178] The memory 902 may be used to store computer programs and data. The computer programs stored in the memory 902 include instructions that can be executed in the processor. The computer programs may constitute various functional modules. The processor 901 executes various functional applications and data processing by calling the computer programs stored in the memory 902.

[0179] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0180] In the solution of the present invention, an automated detection question-answering interactive system is provided to reduce labor costs.

[0181] In the solution of the present invention, according to the time statistics results, a quick answer search is performed according to the questions raised by the client user to achieve a quick question-answer interaction.

[0182] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0183] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0184] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0185] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0186] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A distributed multimodal online customer service intelligent interaction method, characterized in that: The method includes: Set up the infrastructure layer to complete the construction of the infrastructure; After acquiring data at the infrastructure layer, online knowledge base management is performed; Based on the online knowledge base management, set up course information management; Perform time statistics on all historical question and answer information; According to the time statistics, quickly search for answers to questions raised by client users; Make a detailed kanban presentation.

2. The intelligent interaction method of distributed multimodal online customer service according to claim 1, characterized in that: The infrastructure layer is set up to complete the construction of the infrastructure, specifically including: Set up large model servers and application servers; Set up the K8S containerized platform; Set up the MySQL database; Set up a low-code platform.

3. The intelligent interaction method of distributed multimodal online customer service according to claim 1, characterized in that: After acquiring data at the infrastructure layer, online knowledge base management is performed, specifically including: Setting up knowledge base management includes creating, viewing, searching, logging off and clearing functions; Under the New Function, click the Create QA button, a pop-up window will pop up, enter the question and answer fields, and click OK to save; Under the View function, click the View Details button to pop up a pop-up window that displays field information such as the question ID and data source, which cannot be edited; Under the search function, enter the conditions in the problem description and date box, click the search button, the system will search according to the conditions, and the page will display the search results; Under the Clear function, click the Clear button to clear the search conditions in the condition input box; under the Offline function, click the Offline button and a pop-up window will appear, asking if you are sure to go offline. Click OK and the data will enter the pending review interface.

4. The intelligent interaction method of distributed multimodal online customer service according to claim 1, characterized in that: Based on the online knowledge base management, course information management is set up, specifically including: The page includes new creation, deletion, editing, and search functions; To create a new function, you need to fill in the course id and course name; Click the Edit button in the Operation column to enter the editing interface and modify the original course information; Click the Delete button to delete the data; The page supports fuzzy search based on course name.

5. The intelligent interaction method of distributed multimodal online customer service according to claim 4, characterized in that: The time statistics of all historical question and answer information specifically include: Set the statistical categories to monthly statistics, quarterly statistics and yearly statistics; Statistics are collected based on the number of fuzzy searches and exports of the followed users; Each statistic includes the total number of questions and answers, effectiveness, like rate, and failure rate fields.

6. The intelligent interaction method of distributed multimodal online customer service according to claim 5, characterized in that: The method of performing a quick answer search according to the time statistics results and the questions raised by the client user specifically includes: The vocabulary management page includes functions such as search by sensitive word type, fuzzy search by sensitive word, create, edit, and delete; Search by sensitive word type to form the first number; Fuzzy search based on sensitive words to generate a second number; Get the questions raised by the client user, perform vocabulary extraction, and determine whether there are sensitive words; If there is a sensitive word, the first calculation formula is used to calculate the current most matching sensitive word feedback number; Feedback the preset feedback answer based on the current most matching sensitive word feedback number; Get all current matching words, get the time group where the current word is located, the time minute is preset, and the time group is divided into one of monthly statistics, quarterly statistics and yearly statistics; Performing validity calculation using a second calculation formula according to the fuzzy search and the number of derivation times; Based on all current matching words, use the big model to obtain the corresponding optimal feedback list under the current knowledge base and course; When the effectiveness does not exceed the preset value, a comprehensive index is calculated using a third calculation formula according to the optimal feedback list and the total number of answers, effectiveness rate, like rate and failure rate fields corresponding to the answers in the optimal feedback list; When the effectiveness exceeds a preset value, the associated feedback of the answer associated with the current search in the exported data is incorporated into the optimal feedback list; And calculate the comprehensive index using the third calculation formula based on the total number of answers, efficiency, like rate and failure rate fields; The result of the comprehensive index is used as the first feedback result. When the effectiveness exceeds a preset value, the answer with the highest comprehensive index in the exported data associated with the current search is used as the second feedback result. The first calculation formula is: B=k1×S1+k2×S2 Wherein, B is the current most matched sensitive word feedback number, k1 is the first preset coefficient, k2 is the second preset coefficient, S1 is the first number, and S2 is the second number; The second calculation formula is: Y=D / M Among them, Y is validity, D is the number of exports, and M is fuzzy search; The third calculation formula is: ZZ=DZ×xl×(1-sb) 3 ×(1+dianz) Among them, ZZ is the comprehensive index, DZ is the total number of answers, xl is the effective rate, dianz is the like rate, and sb is the failure rate.

7. The intelligent interaction method of distributed multimodal online customer service according to claim 1, characterized in that: The detailed kanban display specifically includes: Set up four follow modules, namely: follow user selection, data statistics, line chart and data report; You can choose to view all data or the data of a certain followed user as the selected user; Set data statistics including total number of questions and answers, total number of questions, number of visits, total number of likes and dislikes, efficiency, like rate and failure rate; Count the number of questions and answers and the number of questions every day in the past 5 days and make a line graph; Count the number and proportion of all likes and dislikes as a data report.

8. A distributed multimodal online customer service intelligent interaction system, characterized in that: The system is used to implement the method according to any one of claims 1 to 7, and the system comprises: The infrastructure setting module is used to set up the infrastructure layer and complete the construction of the infrastructure; The knowledge base management module is used to perform online knowledge base management after acquiring data at the infrastructure layer; The course management module is used to set up course information management based on the online knowledge base management; Time statistics module, used to perform time statistics on all historical question and answer information; Intelligent recommendation module, used to quickly retrieve answers to questions raised by client users based on time statistics results; The kanban display module is used for detailed kanban display.

9. A computer-readable storage medium storing computer program instructions, characterized in that: The computer program instructions implement the method according to any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1-7.

Images