Method for processing user feedback information, electronic equipment and storage medium

Abstract

The invention discloses a method for processing user feedback information, electronic equipment and a storage medium, and belongs to the field of computers. The method comprises the following steps: acquiring feedback information of a target user; based on the feedback information of the target user, determining a question category through a large language model; under the condition that the problem category comprises a technical fault category, first diagnosis data corresponding to the target user is acquired, the first diagnosis data is diagnosis data associated with the technical fault category, and the first diagnosis data is used for determining a first processing result for the feedback information; under the condition that the question category comprises a non-technical fault category, querying a question answer matched with the feedback information from a target knowledge base in an RAG mode; and under the condition that the question answer matched with the feedback information is not queried in the target knowledge base, second diagnosis data corresponding to the target user are obtained, the second diagnosis data are diagnosis data associated with the non-technical fault class, and the second diagnosis data are used for determining a second processing result for the feedback information.

Description

Technical Field

[0001] The present application belongs to the field of computers, and specifically relates to a method for processing user feedback information, an electronic device, and a storage medium. Background Art

[0002] After the product is launched, users will feedback a large number of questions through various channels, such as questions about the use of a certain function of the product. After receiving these questions, it is generally necessary to find the answers to these questions first, and then use these answers to respond to these questions.

[0003] Related technologies usually use keyword matching to find answers to questions. However, in many cases, the answer to the question cannot be found by keyword matching, and a lot of additional manual intervention is still required, resulting in low efficiency in processing feedback information. Summary of the invention

[0004] The embodiments of the present application provide a method, electronic device and storage medium for processing user feedback information, which can solve the problem that when the relevant technology cannot find the answer to the question through keyword matching, a large amount of additional manual intervention is still required, resulting in low efficiency in processing the feedback information.

[0005] In a first aspect, an embodiment of the present application provides a method for processing user feedback information, including: Get feedback from target users; Based on the feedback information of the target user, determining the problem category through a large language model, wherein the problem category includes one of a technical failure category and a non-technical failure category; In a case where the problem category includes a technical failure category, obtaining first diagnostic data corresponding to the target user, where the first diagnostic data is diagnostic data associated with the technical failure category, and the first diagnostic data is used to determine a first processing result of the feedback information of the target user; In the case where the question category includes a non-technical fault category, an answer to the question matching the feedback information is queried from the target knowledge base through a retrieval enhanced generation method; in the case where no answer to the question matching the feedback information is found in the target knowledge base, second diagnostic data corresponding to the target user is obtained, and the second diagnostic data is diagnostic data associated with the non-technical fault category; the second diagnostic data is used to determine a second processing result of the feedback information for the target user.

[0006] In a second aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect or the second aspect are implemented.

[0007] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed, the steps of the method described in the first aspect or the second aspect are implemented.

[0008] In a fourth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the method described in the first aspect or the second aspect.

[0009] At least one of the above technical solutions provided in the embodiments of the present application can achieve the following technical effects: In an embodiment of the present application, feedback information of a target user is obtained; based on the feedback information of the target user, a question category is determined through a large language model, and the question category includes one of a technical failure category and a non-technical failure category; in a case where the question category includes a technical failure category, first diagnostic data corresponding to the target user is obtained, and the first diagnostic data is diagnostic data associated with the technical failure category; the first diagnostic data is used to determine a first processing result for the feedback information of the target user; in a case where the question category includes a non-technical failure category, a question answer matching the feedback information is queried from a target knowledge base through a retrieval enhanced generation method; in a case where a question answer matching the feedback information is not queried in the target knowledge base, second diagnostic data corresponding to the target user is obtained, and the second diagnostic data is diagnostic data associated with the non-technical failure category; the second diagnostic data is used to determine a second processing result for the feedback information of the target user. In this way, in the process of processing feedback information, the question category is first determined through the large language model, so that the accurate question category can be obtained, thereby ensuring that the processing of the feedback information is more targeted; at the same time, by obtaining different diagnostic data (first diagnostic data or second diagnostic data) under different categories, and using different diagnostic data to obtain different processing results for the feedback information, the efficiency of processing the feedback information can be improved, and the problem of low efficiency in processing the feedback information can be solved when the relevant technology cannot find the answer to the question through keyword matching, and a large amount of additional manual intervention is still required. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 is a flow chart of a method for processing user feedback information provided by an embodiment of the present application; Figure 2 is a flowchart of another method for processing user feedback information provided by an embodiment of the present application; Figure 3 It is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0013] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0014] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] When users use products (such as various application products), they will encounter various problems. Some problems are that they do not understand how to use a certain function, some are problems with product implementation, and some are questions about data such as transactions or assets. At this time, they will be submitted to the internal platform through various means (such as contacting customer service).

[0016] In this solution, by introducing the artificial intelligence Large Language Model (LLM) and Retrieval-augmented Generation (RAG) technology, these manual judgment processes are automated. On the one hand, artificial intelligence can be used to quickly determine the category of the problem and guide it to the corresponding processing role. On the other hand, RAG can be used to find the answer to the corresponding question in the existing knowledge base. If there is an answer, it can be directly replied, eliminating the process of manual reply. Moreover, after determining the category of the problem, by obtaining different diagnostic data (first diagnostic data or second diagnostic data) under different categories, and using different diagnostic data to obtain different processing results for feedback information, it can greatly reduce the dependence on manual processing and improve the processing efficiency of feedback information.

[0017] The method for processing user feedback information provided in the embodiment of the present application can be executed by an electronic device, wherein the electronic device can be a server device, the server device can include one or more servers, and the server can be deployed in the cloud.

[0018] Figure 1 is a flow chart of a method for processing user feedback information provided by an embodiment of the present application. Figure 1 The method for processing user feedback information provided in the embodiment of the present application includes: Step 110, obtaining feedback information from target users; In an embodiment of the present application, the target user's feedback information may be feedback information for a product, and the feedback information may be in the form of questions, such as questions reported after using the product. The product may be a target application, such as a stock-related application. In an embodiment, the feedback information is information reported by the target user after using the target application.

[0019] Step 120, based on the feedback information of the target user, determine the problem category through a large language model, wherein the problem category includes one of a technical failure category and a non-technical failure category; In the embodiment of the present application, the large language model can be a model built into the electronic device, or it can be a large language model on the cloud device. If the large language model is a large language model on the cloud device, the electronic device executing the method provided in the embodiment of the present application can call the large language model on the cloud device.

[0020] In one embodiment of the present application, the step 120 of determining the question category through a large language model based on the feedback information of the target user may include: identifying the target image through a large language model to obtain the image type and image content of the target image; the image type includes mobile phone images and computer images; determining the usage scenario based on the image content of the target image; determining the question category through the large language model based on the image type, the usage scenario and the image content. In this way, the question category can be determined more accurately through the large language model combined with the image type, the image usage scenario and the image content, thereby ensuring that the subsequent problem processing is more targeted and accurate.

[0021] The usage scenarios in the embodiments of the present application may include: at least one of an information scenario, an asset scenario, and a market scenario. The usage scenario determined based on the image content of the target image may involve one of an information scenario, an asset scenario, and a market scenario. Among them, the usage scenario in the embodiments of the present application refers to the usage scenario of the product. Taking the product as the target application as an example, the usage scenario may represent the usage scenario for the target application, that is, the specific scenario of using the target application.

[0022] In the embodiment of the present application, the problem categories may include technical failure categories and non-technical failure categories. The non-technical failure categories include at least one of user experience categories and function usage question categories. The user experience category may include some areas for improvement raised by the target user after using the product. The function usage question category may include questions raised by the target user because they do not know how to use a certain function.

[0023] Step 130: when the problem category includes a technical failure category, obtaining first diagnostic data corresponding to the target user, the first diagnostic data being diagnostic data associated with the technical failure category; the first diagnostic data being used to determine a first processing result of the feedback information of the target user; In an embodiment of the present application, the first diagnostic data may include at least one of client diagnostic data, server diagnostic data, and infrastructure diagnostic data. Among them, the infrastructure diagnostic data may include network-related diagnostic data, etc. The first diagnostic data may be obtained by querying logs. The client diagnostic data may include diagnostic data related to client performance. The server diagnostic data may include diagnostic data related to server performance.

[0024] In the embodiment of the present application, after obtaining the first diagnostic data, the electronic device may obtain a first processing result of the feedback information of the target user based on the first diagnostic data, and reply to the target user using the first processing result.

[0025] It should be understood that in an embodiment of the present application, the electronic device can also analyze the target department for processing feedback information based on the feedback information of the target user through a large language model. In this case, after obtaining the first diagnostic data, the electronic device can transmit the first diagnostic data to the target device corresponding to the target department. Furthermore, the target device corresponding to the target department can obtain the first processing result based on the first diagnostic data, and reply to the target user using the first processing result. In this way, by having the target device of the corresponding department reply to the target user using the first processing result obtained based on the first diagnostic data, replies to unfamiliar situations can be avoided and the accuracy of the reply can be improved.

[0026] The target department includes one of a first department associated with a client, a second department associated with a server, and a third department associated with an infrastructure. When the target department includes the first department, the first diagnostic data includes client diagnostic data; when the target department includes the second department, the first diagnostic data includes server diagnostic data; when the target department includes the third department, the first diagnostic data includes infrastructure diagnostic data. In this way, by obtaining specific diagnostic data corresponding to a department, an accurate first processing result can be obtained.

[0027] Step 140, when the question category includes a non-technical fault category, querying the answer to the question matching the feedback information from the target knowledge base through a retrieval enhancement generation method; In the embodiment of the present application, the large language model and the search enhancement generation can be used in combination. The target knowledge base can store various question answers matching various feedback information. Among them, the various feedback information can be feedback information from different users.

[0028] It should be understood that in the embodiment of the present application, when the target knowledge base finds the answer to the question that matches the feedback information, the electronic device can reply to the target user using the answer to the question. In this way, for the questions already in the target knowledge base, the large language model can better match the questions and answers, thereby accurately returning the answers to the questions and reducing the manual workload.

[0029] Step 150, when no answer to the question matching the feedback information is found in the target knowledge base, obtain second diagnostic data corresponding to the target user, wherein the second diagnostic data is diagnostic data associated with the non-technical fault class; the second diagnostic data is used to determine a second processing result for the feedback information of the target user.

[0030] Among them, the second diagnostic data can be obtained in the form of a log. After obtaining the second diagnostic data, the electronic device can obtain a second processing result based on the second diagnostic data. Furthermore, the second processing result obtained based on the second diagnostic data can be used to reply to the target user, and the second processing result can also be added to the target knowledge base as an answer to the question matched by the feedback information. In this way, for non-technical failure problems, after the processing is completed, they can be added to the target knowledge base again to ensure the continuous updating of the target knowledge base to cope with feedback from other users in the future.

[0031] In an embodiment of the present application, feedback information of a target user is obtained; based on the feedback information of the target user, a problem category is determined through a large language model, and the problem category includes one of a technical failure category and a non-technical failure category; when the problem category includes the technical failure category, first diagnostic data corresponding to the target user is obtained, and the first diagnostic data is diagnostic data associated with the technical failure category; the first diagnostic data is used to determine a first processing result for the feedback information of the target user; when the problem category includes the non-technical failure category, a question answer matching the feedback information is queried from a target knowledge base through a RAG method; when no question answer matching the feedback information is found in the target knowledge base, second diagnostic data corresponding to the target user is obtained, and the second diagnostic data is diagnostic data associated with the non-technical failure category; the second diagnostic data is used to determine a second processing result for the feedback information of the target user. In this way, in the process of processing feedback information, the question category is first determined through the large language model, so that the accurate question category can be obtained, thereby ensuring that the processing of the feedback information is more targeted; at the same time, by obtaining different diagnostic data (first diagnostic data or second diagnostic data) under different categories, and using different diagnostic data to obtain different processing results for the feedback information, the efficiency of processing the feedback information can be improved, and the problem of low efficiency in processing the feedback information can be solved when the relevant technology cannot find the answer to the question through keyword matching, and a large amount of additional manual intervention is still required.

[0032] Figure 2 It is a flowchart of a method for processing user feedback information provided by an embodiment of the present application. Figure 2 The method shown can be executed by an electronic device. Figure 2 The method for processing user feedback information provided in the embodiment of the present application includes: Step 210, obtaining feedback information from a target user, wherein the feedback information includes a target picture; The feedback information may be information fed back by the target user after using the target application. The target image may be an image of the product being used. Taking the product as the target application as an example, the target image may be an image displayed during the use of the target application.

[0033] Step 220, identifying the target image through a large language model to obtain the image type and image content of the target image; the image type includes mobile phone images and computer images; Among them, the product in the embodiment of the present application can be a software product. During the development process, software products for computer terminals and software products for mobile terminals can be developed. By identifying the target image through a large language model, it can be obtained whether the target image belongs to the mobile terminal image of the software product or the computer terminal image of the software product.

[0034] Step 230, determining a usage scenario based on the image content of the target image; The usage scenario includes one of an information scenario, an asset scenario, and a market scenario. The usage scenario in the embodiment of the present application refers to the usage scenario of the product. Taking the product as the target application as an example, the usage scenario can represent the usage scenario for the target application, that is, the specific scenario of using the target application.

[0035] In the embodiment of the present application, the usage scenario can also be determined based on the image content and image type of the target image.

[0036] Step 240, determining a problem category through the large language model based on the image type, the usage scenario, and the image content, wherein the problem category includes one of a technical failure category and a non-technical failure category; The non-technical failure category includes at least one of a user experience category and a function usage question category.

[0037] Step 250: when the problem category includes a technical failure category, obtaining first diagnostic data corresponding to the target user, the first diagnostic data being diagnostic data associated with the technical failure category; the first diagnostic data being used to determine a first processing result of the feedback information of the target user; Step 255, replying to the target user using the first processing result obtained based on the first diagnostic data; In one embodiment of the present application, the electronic device may acquire first diagnostic data, and obtain a first processing result based on the first diagnostic data. Furthermore, the electronic device may reply to the target customer using the first processing result.

[0038] In another embodiment of the present application, after obtaining the feedback information of the target user, the electronic device can determine the target department for processing the feedback information based on the feedback information of the target user, and then send the first diagnostic data to the target device corresponding to the target department. Subsequently, the target device corresponding to the target department can reply to the target user using the first processing result obtained based on the first diagnostic data.

[0039] The target department includes one of a first department associated with a client, a second department associated with a server, and a third department associated with an infrastructure. When the target department includes the first department, the first diagnostic data includes client diagnostic data; when the target department includes the second department, the first diagnostic data includes server diagnostic data; when the target department includes the third department, the first diagnostic data includes infrastructure diagnostic data.

[0040] Step 260, when the question category includes a non-technical fault category, querying the target knowledge base for a question answer matching the feedback information by using a retrieval enhancement generation method; Step 270: When an answer to a question matching the feedback information is found in the target knowledge base, the target user is replied with the answer to the question.

[0041] Step 280: if no answer to the question matching the feedback information is found in the target knowledge base, obtain second diagnostic data corresponding to the target user, where the second diagnostic data is diagnostic data associated with the non-technical fault class; the second diagnostic data is used to determine a second processing result for the feedback information of the target user; In this step, the electronic device may obtain a second processing result based on the second diagnostic data.

[0042] Step 290: reply the target user using the second processing result obtained based on the second diagnostic data, and add the second processing result as an answer to the question matched by the feedback information to the target knowledge base.

[0043] In the embodiment of the present application, the electronic device can obtain the second diagnostic data, and obtain the second processing result based on the second diagnostic data. Furthermore, the electronic device can reply to the target customer using the second processing result.

[0044] In another embodiment of the present application, after obtaining the feedback information of the target user, the electronic device can determine the target department for processing the feedback information based on the feedback information of the target user, and then send the second diagnostic data to the target device corresponding to the target department. Subsequently, the target device corresponding to the target department can reply to the target user using the second processing result obtained based on the second diagnostic data.

[0045] Wherein, the target department includes one of a first department associated with a client, a second department associated with a server, and a third department associated with an infrastructure. When the target department includes the first department, the second diagnostic data includes client diagnostic data; when the target department includes the second department, the second diagnostic data includes server diagnostic data; when the target department includes the third department, the second diagnostic data includes infrastructure diagnostic data. Wherein, the infrastructure diagnostic data may include diagnostic data related to the network, etc. The second diagnostic data may be obtained by querying logs.

[0046] It should be understood that in the embodiments of the present application, Figure 2 and Figure 1 The same or corresponding steps in the above can be referenced to each other. Figure 2 Zhongyu Figure 1 The same or corresponding steps can be referred to in the previous article Figure 1 's discussion.

[0047] In an embodiment of the present application, in the process of processing feedback information, the question category is first determined through a large language model, and an accurate question category can be obtained, thereby ensuring that the processing of the feedback information is more targeted; at the same time, by obtaining different diagnostic data (first diagnostic data or second diagnostic data) under different categories, and using different diagnostic data to obtain different processing results for the feedback information, the efficiency of processing the feedback information can be improved, and the problem of low efficiency in processing the feedback information caused by the fact that the relevant technology cannot find the answer to the question through keyword matching, and a large amount of additional manual intervention is still required is solved.

[0048] Moreover, the embodiment of the present application combines a large language model with user feedback processing. Through the text analysis capability of the large language model, it can more accurately analyze the type of feedback content and the exact department that needs to handle the problem, as well as the scenario in which the problem occurs. Relevant diagnostic data can be prepared in advance for machines or engineers to quickly check, and knowledge base retrieval can be performed through the RAG method, which can have a higher hit rate than traditional keyword matching and other methods.

[0049] The following is a further explanation of the method for processing user feedback information provided in the embodiment of the present application.

[0050] The method for processing user feedback information provided in the embodiment of the present application may include the following parts: using a multimodal large language model to identify and classify questions, and using RAG to retrieve questions from a knowledge base, etc.

[0051] The specific process of the method for processing user feedback information provided in the embodiment of the present application can be as follows: After receiving the target user's feedback information, first determine whether it contains the target image. If the feedback information contains the target image, use the multimodal LLM to identify the image content, and submit the identified content and the text content of the user feedback to the LLM. The LLM also classifies the user feedback information, such as whether the problem belongs to the technical failure category, user experience category, or function usage question category. Among them, the user experience category and function usage question category belong to the non-technical failure category.

[0052] At the same time, LLM determines which department will handle the problem based on the content and pictures of user feedback. For example, if a user reports a problem related to a mobile app, it will be handed over to the first department associated with the client (client team). After determining the processing department, diagnostic data associated with the user (including client diagnostic data) is obtained at the same time, which facilitates subsequent rapid judgment of the problem by at least one of the machine and the engineer.

[0053] For questions about function usage and user experience, there are usually corresponding solutions in the target knowledge base that has been prepared in advance. For usage questions, previous feedback answers and documents in the help center can be used to match and obtain answers from the target knowledge base through RAG, and the answers can be given to the target users.

[0054] For non-technical failure issues (including questions about function usage and user experience), after processing, the answers to the issues can be added to the target knowledge base again to respond to future feedback from other users.

[0055] In this way, the multimodal large language model can directly analyze the text content of user feedback and summarize the functional module to which the user wants to respond. If the user's feedback is an image, the multimodal large language model can also be used to identify the image content and summarize the specific content and the scene where the problem occurred. Moreover, for existing questions in the knowledge base, the large language model can better match questions and answers, thereby accurately returning answers to questions and reducing manual workload.

[0056] The embodiment of the present invention combines a large language model with user feedback processing. Through the text analysis capability of the large language model, it can more accurately analyze the category of feedback content and the exact department that needs to handle the problem, as well as the scenario where the problem occurs. Relevant diagnostic data can be prepared in advance for engineers or machines to quickly check, and knowledge base retrieval can be performed through the RAG method, which can have a higher hit rate than traditional keyword matching methods.

[0057] In the embodiment of the present invention, Python can be used as a programming language. In terms of machine learning and large language models, Python has many mature libraries available for use, which can quickly complete the call of large language models and RAG-related development work.

[0058] Figure 3 is a structural block diagram of an electronic device provided in an embodiment of the present application. Figure 3 As shown, the embodiment of the present application further provides an electronic device 300. The electronic device 300 includes: a processor 310 and a memory 320, the memory 320 stores programs or instructions, and when the programs or instructions are executed by the processor 310, the steps of any of the methods described above are implemented. For example, when the program is executed by the processor 310, the following process is implemented: obtaining feedback information of the target user; based on the feedback information of the target user, determining the problem category through a large language model, the problem category includes one of a technical failure category and a non-technical failure category; when the problem category includes a technical failure category, obtaining first diagnostic data corresponding to the target user, the first diagnostic data is diagnostic data associated with the technical failure category; the first diagnostic data is used to determine a first processing result for the feedback information of the target user; when the problem category includes a non-technical failure category, querying the target knowledge base for a question answer matching the feedback information through a retrieval enhancement generation method; when no question answer matching the feedback information is found in the target knowledge base, obtaining second diagnostic data corresponding to the target user, the second diagnostic data is diagnostic data associated with the non-technical failure category; the second diagnostic data is used to determine a second processing result for the feedback information of the target user. In this way, in the process of processing feedback information, the question category is first determined through the large language model, so that the accurate question category can be obtained, thereby ensuring that the processing of the feedback information is more targeted; at the same time, by obtaining different diagnostic data (first diagnostic data or second diagnostic data) under different categories, and using different diagnostic data to obtain different processing results for the feedback information, the efficiency of processing the feedback information can be improved, and the problem of low efficiency in processing the feedback information can be solved when the relevant technology cannot find the answer to the question through keyword matching, and a large amount of additional manual intervention is still required.

[0059] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of each embodiment of the method for processing user feedback information are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0060] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0061] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0062] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0063] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0064] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0065] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for processing user feedback information, characterized in that: include: Get feedback from target users; Based on the feedback information of the target user, determining the problem category through a large language model, wherein the problem category includes one of a technical failure category and a non-technical failure category; In the case where the problem category includes a technical failure category, obtaining first diagnostic data corresponding to the target user, the first diagnostic data being diagnostic data associated with the technical failure category; the first diagnostic data being used to determine a first processing result of the feedback information of the target user; In the case where the question category includes a non-technical fault category, an answer to the question matching the feedback information is queried from the target knowledge base through a retrieval enhanced generation method; in the case where no answer to the question matching the feedback information is found in the target knowledge base, second diagnostic data corresponding to the target user is obtained, and the second diagnostic data is diagnostic data associated with the non-technical fault category; the second diagnostic data is used to determine a second processing result of the feedback information for the target user.

2. The method according to claim 1, characterized in that: The feedback information includes a target image; and determining the question category through a large language model based on the feedback information of the target user includes: The target image is identified by a large language model to obtain the image type and image content of the target image; the image type includes mobile phone images and computer images; Determining a usage scenario based on the image content of the target image; Based on the image type, the usage scenario and the image content, the question category is determined by the large language model.

3. The method according to claim 2, characterized in that The feedback information is information fed back by the target user after using the target application, and the usage scenario includes one of an information scenario, an asset scenario, and a market scenario.

4. The method according to any one of claims 1 to 3, characterized in that: The non-technical failure category includes: at least one of a user experience category and a function usage question category; the method further includes: When an answer to a question matching the feedback information is found in the target knowledge base, the target user is replied with the answer to the question.

5. The method according to any one of claims 1 to 3, characterized in that: After obtaining the feedback information of the target user, the method further includes: Based on the feedback information of the target user, determining a target department for processing the feedback information; The target device corresponding to the target department replies to the target user using the first processing result obtained based on the first diagnostic data.

6. The method according to claim 5, characterized in that The target department includes one of a first department associated with the client, a second department associated with the server, and a third department associated with the infrastructure; In the case where the target department includes the first department, the first diagnostic data includes client diagnostic data; In the case where the target department includes the second department, the first diagnostic data includes server-side diagnostic data; In the case where the target department includes a third department, the first diagnostic data includes infrastructure diagnostic data.

7. The method according to claim 1, characterized in that After acquiring the second diagnostic data corresponding to the target user, the method further includes: replying to the target user using the second processing result obtained based on the second diagnostic data; The second processing result is used as the answer to the question matched by the feedback information and added to the target knowledge base.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a program or instruction running on the processor, and when the program or instruction is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

9. A storage medium, characterized in that: The medium stores a program or an instruction, and when the program or the instruction is executed, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images