User report processing method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a user declaration processing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of determining a fault type of user declaration information; calling a corresponding interface, and obtaining signaling information corresponding to the fault type; and inputting the signaling information into a preset signaling analysis large model, and using a prompt project to guide the signaling analysis large model to define and position the signaling information for the user report information, thereby obtaining problem signaling and problems existing in the problem signaling. By adopting the method, the automation of user declaration processing can be realized, and the problem signaling and the problem existing in the problem signaling are more accurate by analyzing the large model through the prompt project guide signaling.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a user declaration processing method, apparatus, computer equipment, computer-readable storage medium and computer program product. Background Art

[0002] With the development of technology, users have an increasing demand for the quality of communication services. User declarations are user complaints. For example, users complain that their calls are automatically disconnected after about one minute. For user declarations (complaints), the current method is manual analysis and processing, which is inefficient. Summary of the invention

[0003] Based on this, the present invention provides a user declaration processing method, apparatus, computer equipment, computer-readable storage medium and computer program product to realize automated processing of user declarations.

[0004] In a first aspect, the present application provides a method for processing user claims, comprising:

[0005] Determine the fault type of the user's reported information;

[0006] Call the corresponding interface to obtain the signaling information corresponding to the fault type;

[0007] Input the signaling information into a preset large model, and use a prompting project to guide the large model to delimit and locate the signaling information according to the user declaration information;

[0008] Based on the output of the large model, the problem signaling corresponding to the user declaration information and the problems existing in the problem signaling are obtained.

[0009] In one embodiment, the prompting project includes a BROKE prompt word framework; and using the prompting project to guide the large model includes at least one of the following:

[0010] If the initial prompt content under the BROKE prompt word framework has missing content, the missing prompt content is filled in by interacting with the outside to obtain the target prompt content and input it into the large model;

[0011] Determine a signaling analysis prompt template corresponding to the fault type from a plurality of preset prompt engineering templates, and input the signaling analysis prompt template into the large model; the plurality of prompt engineering templates are set based on a BROKE prompt word framework;

[0012] According to the user declaration information and the preset knowledge base, signaling analysis prompt content for the fault type is generated based on the BROKE prompt word framework and input into the large model as the target prompt content.

[0013] In one of the embodiments, determining the signaling analysis prompt template corresponding to the fault type from a plurality of preset prompt engineering templates includes:

[0014] Based on the signaling analysis rule corresponding to the fault type, a signaling analysis prompt template corresponding to the fault type is determined from a plurality of preset prompt engineering templates.

[0015] In one embodiment, the method further comprises:

[0016] When it is determined that the signaling analysis prompt template does not contain the content of the signaling analysis rule, the signaling analysis rule is input into the large model.

[0017] In one embodiment, before determining the fault type of the user's reported information, the method further includes:

[0018] Determine the processing flow corresponding to the user's declaration based on the user's declaration information and the business process information in the preset knowledge base;

[0019] The calling of the corresponding interface to obtain signaling information corresponding to the fault type includes:

[0020] Determining at least one corresponding data interface based on the processing flow;

[0021] The data interface is called to obtain signaling information corresponding to the fault type.

[0022] In one embodiment, determining the fault type of the user's reported information includes:

[0023] The user declaration information is input into a preset big model, and the fault type of the user declaration information is identified through the semantics of the big model.

[0024] In one embodiment, the method further comprises:

[0025] In the case where the fault type is a preset fault type, querying the setting information of the user terminal corresponding to the user declaration information, and checking whether the setting information of the user terminal is passed;

[0026] The calling of the corresponding interface to obtain signaling information corresponding to the fault type includes:

[0027] In the case where the fault type is a preset fault type and the check result of the setting information of the user terminal is passed, a corresponding interface is called to obtain signaling information corresponding to the fault type.

[0028] In one embodiment, after obtaining the problem signaling corresponding to the user declaration information and the problem existing in the problem signaling, the method further includes:

[0029] According to the problem signaling and the problem existing in the problem signaling, combined with the processing rule information in the preset knowledge base, corresponding processing suggestion information is obtained, and the processing suggestion information is sent to the corresponding entity.

[0030] In a second aspect, the present application also provides a user declaration processing device, comprising:

[0031] An identification module, used to determine the fault type of the user's reported information;

[0032] An acquisition module, used to call a corresponding interface to obtain signaling information corresponding to the fault type;

[0033] An analysis module, used for inputting the signaling information into a preset large model, and using a prompting project to guide the large model to delimit and locate the signaling information according to the user declaration information;

[0034] A determination module is used to obtain the problem signaling corresponding to the user declaration information and the problems existing in the problem signaling based on the output of the large model.

[0035] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in any one of the first aspects are implemented.

[0037] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0038] The above-mentioned user report processing method, apparatus, computer equipment, computer-readable storage medium and computer program product can realize the automation of user report processing by determining the fault type of the user report information, obtaining the signaling information corresponding to the fault type, and using the big model to analyze the corresponding signaling information to determine the problem signaling; and also guide the big model through prompt engineering, so that the output of the model is more in line with the needs of user report processing, and the problem signaling and the problems existing in the problem signaling are obtained more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A flowchart of a method for processing user claims in one embodiment;

[0041] Figure 2 is a schematic diagram of a business processing flow in a knowledge base in one embodiment;

[0042] Figure 3 Another flowchart of a method for processing user claims in one embodiment;

[0043] Figure 4 It is another flowchart of a method for processing user declarations in one embodiment;

[0044] Figure 5 It is another flowchart of a method for processing user declarations in one embodiment;

[0045] Figure 6 It is a structural block diagram of a user declaration processing device in one embodiment;

[0046] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] In one embodiment, Figure 1As shown, a method for processing user declarations is provided. This embodiment uses the method to be applied to a pre-set intelligent agent Agent in a communication network as an example. It can be understood that the method can also be applied to a server or terminal in a communication network, in which an intelligent agent Agent is pre-set (agent, Agent, is a core concept in the field of artificial intelligence, referring to any entity that can autonomously process information, make decisions and take actions to achieve specific goals). The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. It can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. It can also be applied to other systems, such as a telephone self-service system. In this embodiment, the method includes steps S101 to S104:

[0049] Step S101: Determine the fault type of the user's reported information.

[0050] Exemplarily, user declaration information is user complaints, such as user complaints that each call is automatically disconnected after about a minute. In a specific embodiment, a large model (such as a semantic recognition model based on a general large model and trained using data related to user declaration information) can be used for semantic recognition to accurately determine the fault encountered by the user. The type of fault reported by the user is a type of fault corresponding to the user's declaration. For example, call failure, Internet failure, etc. Semantic recognition is performed on the user's declaration information to determine the corresponding fault type. For example, the user reported that the call was automatically disconnected after about a minute, and the semantic recognition showed that it was a call failure.

[0051] Step S102: calling a corresponding interface to obtain signaling information corresponding to the fault type.

[0052] Exemplarily, the corresponding interface, that is, the interface for obtaining signaling information corresponding to the aforementioned fault type, may not necessarily correspond to the same interface for different fault types. By calling the corresponding interface, the signaling information corresponding to the fault type can be obtained. Signaling information may be information related to signaling. For example, a signaling file (SignalingFile) refers to a file used to store and process signaling messages in a communication system. Signaling messages are messages used to control, manage, monitor and maintain the communication system during the communication process. They usually contain information about call establishment, removal, modification and other communication-related operations.

[0053] Step S103: input the signaling information into a preset large model, and use the prompt engineering to guide the large model to delimit and locate the signaling information according to the user declaration information.

[0054] The large model in the embodiment of the present application can be a machine learning model with a large number of parameters and complex structure that can be used for signaling analysis, which can find out the signaling with abnormalities or problems in the signaling information and give the problems of the related signaling. Delimitation and positioning, that is, finding out which signaling has what kind of problem, can also refer to determining the specific location where the problem or fault occurs. It is an important step in fault handling. Delimitation focuses more on quickly isolating the scope of the problem, while positioning is to further analyze and determine the specific cause of the problem after determining the scope of the problem. In the embodiment of the present application, delimitation and positioning can be understood as finding out which signaling has what kind of problem. Prompt engineering is also called prompt engineering, Prompt, Prompt Engineering, etc. It is a natural language processing (NLP) technology that aims to guide AI models to generate more accurate and useful answers through carefully designed prompts. Prompt design is the key to effective interaction using language models (especially large pre-trained language models, etc.). A good prompt can guide the model to output high-quality, relevant and expected content. Exemplarily, the prompt project may include: Instruction (which is the most critical part of the prompt, directly telling the model the specific task the user wants to perform), Input Data (the specific information that the model needs to process, which should be clearly identified in the prompt), Context (which provides the model with the environmental information or additional details required to perform the task, which helps the model better understand the environment or context in which the task is located), Output Indicator (which defines the expected type or format of the model output, and helps subsequent modules automatically parse the output results of the model), etc. Further, the prompt project may include a variety of different prompt word frameworks, such as APE (Action Purpose Expect) framework, BROKE framework, CHAT framework, CRISPE framework, etc. The signaling information is input into a preset large model, and the large model is guided by the prompt project, so that the large model's delimitation and positioning of the signaling information for the user's declaration information meets the requirements of the prompt project, that is, meets the actual user declaration situation.

[0055] Step S104: Based on the output of the large model, obtain the problem signaling corresponding to the user's declaration information and the problems existing in the problem signaling.

[0056] Exemplarily, the problem signaling corresponding to the user's declaration information and the problems existing in the problem signaling, that is, the problem signaling related to the user's declaration information and the problems existing in the problem signaling. The signaling information is analyzed by the large model and the corresponding analysis results are output, so that the relevant problem signaling and the problems existing in the problem signaling can be determined.

[0057] In the above user report processing method, the corresponding fault type is obtained through semantic recognition of user report information; the signaling information corresponding to the fault type is obtained, and the corresponding signaling information is analyzed using a large model to determine the problem signaling, thereby realizing the automation of user report processing. In addition, by guiding the large model through prompt engineering, the output of the model is more in line with the needs of user report processing, and the problem signaling and the problems existing in the problem signaling are obtained more accurately.

[0058] In one embodiment, the "prompt project" in the aforementioned embodiment includes a BROKE prompt word framework; the step of "using the prompt project to guide the large model" in the aforementioned embodiment includes at least one of the following: if the initial prompt content under the BROKE prompt word framework is missing content, the missing prompt content is filled in by interacting with the outside to obtain the target prompt content and input it into the large model; a signaling analysis prompt template corresponding to the fault type is determined from multiple preset prompt project templates, and the signaling analysis prompt template is input into the large model; multiple prompt project templates are set based on the BROKE prompt word framework; according to the user declaration information and the preset knowledge base, the signaling analysis prompt content for the fault type is generated based on the BROKE prompt word framework, and is input into the large model as the target prompt content.

[0059] For example, the BROKE prompt word framework, namely the BROKE (Background, Role, Objectives, KeyResults, Evolve) framework, mainly includes the following five framework components: Framework Component 1: Background: This part provides detailed context and background information, explains the environment, historical situation or any relevant prerequisites of the task, and helps the big model understand the source of the demand and the environment in which it is located. Framework Component 2: Role: The role is to define the role that the big model should play in the conversation, such as expert, consultant, analyst, etc., to ensure that the big model's response fits the position and needs of the specified role. Framework Component 3: Objectives: Clearly point out the specific goals you hope to achieve through this conversation. This part clearly states the purpose or goal you hope to achieve through this interaction, which helps the big model focus on the final result you want to achieve and ensure that the content produced directly serves this purpose. Framework Component 4: Key Results: List the specific standards or expected results for measuring whether the goal has been achieved, provide the big model with a way to evaluate the effectiveness of its generated content, and ensure that the solution or information is measurable and targeted. Framework Component 5: Experimental Improvement (Evolve): During the interaction with the large model, if the generated answer is not completely in line with expectations, some strategies need to be adopted to adjust and optimize. This may include improving the input information, pointing out the shortcomings in the answer, or trying to generate results multiple times without changing the original instructions to optimize the output.

[0060] The "BROKE" prompt word framework is a very useful tool in the Prompt project, which helps designers create structured and clear prompts to improve the work efficiency and quality of large models. The preset prompt project template, that is, the template pre-established for guiding the prompt project of the aforementioned large model, is set based on the BROKE prompt word framework, and the template content usually involves Background, Role, Objectives, Key Results, and Evolve. Exemplarily, the preset prompt project template can be designed based on possible different user declaration information, fault types, etc. The prompt project template can include the prompt content itself, and can also include related signaling analysis rules, etc. The preset knowledge base can be a local knowledge base pre-established for the intelligent agent, which can store relevant information related to the user declaration for generating the above-mentioned prompt project and prompt content. For example, the knowledge base can include background information, task objectives, etc. required to generate the prompt content. Of course, the preset knowledge base can also store other relevant information required for better processing user declarations, such as business processing procedures for user declarations.

[0061] In the above embodiment, the prompt project includes a BROKE prompt word framework, and three methods for implementing the prompt project guidance model based on the BROKE prompt word framework are given: (1) Filling: If the initial prompt content under the BROKE prompt word framework is missing, the missing prompt content is filled by interacting with the outside to obtain the target prompt content input into the large model; (2) Selecting: Determine the corresponding signaling analysis prompt template from multiple preset prompt project templates, input the signaling analysis prompt template into the large model, and the prompt project template is set based on the BROKE prompt word framework; (3) Lengthening: According to the user's declaration information and the preset knowledge base, the signaling analysis prompt content for the fault type is dynamically generated based on the BROKE prompt word framework, and is input into the large model as the target prompt content. For example, according to the user's declaration information and the preset knowledge base, the user's declaration processing flow can be planned in advance, and the corresponding prompt content can be dynamically generated according to the processing flow. The prompt content generated in this way can include all parts of Background, Role, Objectives, Key Results, and Evolve, or some or several of them can be omitted. For these three methods, you can choose to use one or a combination. In an exemplary embodiment, for the above three methods, based on a large number of test results, the best method or combination of methods can be used as a method for prompting the project to guide the large model. In another exemplary embodiment, the method corresponding to the processing flow can be pre-defined according to different processing flows, so that when the signaling analysis corresponding to different processing flows is performed, the method corresponding to the flow can be used to guide the large model to perform signaling analysis. In another exemplary embodiment, the three methods can also be used successively to generate prompt content input into the large model, so that the large model can fully understand the intention of the signaling analysis, and the signaling analysis results obtained based on the three methods can be mutually verified, so as to obtain more accurate signaling analysis results.

[0062] In this embodiment, by incorporating the BROKE prompt word framework into the prompt project, the prompt project can guide the large model more accurately, thereby obtaining a more accurate location of the problem signaling; and, by filling in the missing initial prompt content through external interaction, determining the signaling analysis prompt template corresponding to the fault type from multiple preset prompt project templates, and dynamically generating the signaling analysis prompt content for the fault type based on the BROKE prompt word framework according to the user's declaration information and the preset knowledge base, the prompt project can be used to guide the large model in a diversified manner, thereby making the located problem signaling more accurate; at the same time, due to this diversification, there are multiple ways to use the prompt project to guide the large model. When one method fails or the effect is not good (such as due to various unexpected situations), other methods can be selected to implement the prompt project to guide the large model, thereby increasing the fault tolerance of the declaration processing method and improving stability.

[0063] In one embodiment, the "determining a signaling analysis prompt template corresponding to the fault type from multiple preset prompt engineering templates" in the aforementioned embodiment includes: based on the signaling analysis rules corresponding to the fault type, determining the signaling analysis prompt template corresponding to the fault type from multiple preset prompt engineering templates.

[0064] Exemplarily, the signaling analysis rules corresponding to the fault type, that is, the signaling analysis rules for the aforementioned fault type, for example, the signaling analysis rules can be RFC3261, and RFC3261 is a standard document that defines SIP (Session Initiation Protocol) and specifies in detail the rules of SIP message format, transmission mechanism, operation process, etc. If the current fault is a SIP fault, the RFC3261 rule should be used to analyze the signaling, and if it is other faults, other rules should be used to analyze the corresponding signaling information. Therefore, the signaling analysis rules are determined by the fault type, and the signaling analysis prompt template is determined based on the signaling analysis rules, so that the big model can know how to measure the completion of the goal. The signaling analysis prompt template, that is, the prompt template used to prompt the engineering to guide the big model to perform signaling analysis, can be a signaling analysis prompt template pre-set based on different signaling analysis rules. For example, if the fault type is a call fault, the corresponding signaling analysis prompt template may include "Please determine whether there is a problem with the following signaling according to the aforementioned RFC3261 rules."

[0065] In this example, the required signaling analysis prompt template is selected and determined based on the signaling analysis rules corresponding to the fault type, so that the obtained template can be more in line with the actual situation, which will help the subsequent output of the model to better meet the needs of user report processing and make the problem signaling and the problems existing in the problem signaling more accurate.

[0066] In one embodiment, the user declaration processing method in the aforementioned embodiment further includes: when the determined signaling analysis prompt template does not contain the content of the signaling analysis rule, inputting the signaling analysis rule into the large model.

[0067] Exemplarily, the signaling analysis rules can not only help determine the signaling analysis prompt template, but also can be directly input into the big model to provide a prompt project to guide the big model.

[0068] In this example, when the determined signaling analysis prompt template does not contain the content of the signaling analysis rule, the signaling analysis rule is input into the large model, thereby avoiding the situation that the corresponding signaling analysis rule may not exist in the aforementioned prompt template and affect the guidance of the prompt project to the large model. At the same time, by inputting the signaling analysis rule into the large model separately, there is a certain degree of independence between the signaling analysis rule and the aforementioned signaling analysis prompt template, so that the signaling analysis rule can be easily adjusted and modified separately, which increases the flexibility and dynamic variability of the prompt project in guiding the large model.

[0069] In one embodiment, see Figure 5 As shown, before "determining the fault type of the user's report information" in the aforementioned embodiment, the user report processing method also includes: determining the processing flow corresponding to the user's report based on the user's report information and the business process information in the preset knowledge base; in the aforementioned embodiment, "calling the corresponding interface to obtain the signaling information corresponding to the fault type" includes: determining at least one corresponding data interface based on the processing flow; calling the data interface to obtain the signaling information corresponding to the fault type.

[0070] Exemplarily, the preset knowledge base stores relatively complete business process information, which includes the business processes required for different user reports. For example, when the user report information corresponds to an Internet failure, the business processing process information corresponding to the Internet failure can be determined from the complete business process information. Figure 2 A possible specific method is given for reference. Based on the user declaration information and the business process information in the preset knowledge base, the processing flow corresponding to the user declaration can be found. Through the processing flow, the corresponding data interface can be obtained (the data interface corresponding to different processing flows may be different, and the number of corresponding data interfaces may also be different), and the required signaling information can be obtained from the corresponding data interface.

[0071] In this example, considering that different user declaration information may be directed to different problems, and the processing procedures, data interfaces to be called, signaling information to be obtained, etc. corresponding to different problems are not necessarily the same, therefore, before determining the fault type of the user declaration information, the processing procedure corresponding to the user declaration is determined according to the user declaration information and the business process information in the preset knowledge base, and the corresponding data interface is called according to the processing procedure to obtain the signaling information corresponding to the fault type. In this way, it avoids executing unnecessary procedures and obtaining unnecessary signaling information, and can quickly and accurately obtain the required signaling information corresponding to the fault type, which helps to improve the efficiency and accuracy of user declaration processing.

[0072] In one embodiment, "determining the fault type of the user's reported information" in the aforementioned embodiment includes: inputting the user's reported information into a preset big model, and identifying the fault type of the user's reported information through the semantics of the big model.

[0073] For the preset big model, please refer to the relevant part above. Semantic recognition is to identify the meaning of the user's declaration information. For example, semantic recognition can be to extract information from the text and convert it into structured data that can be processed by the computer, so that the computer can understand and parse the meaning in human language. For example, by using the big model to perform semantic recognition on the user's declaration information, the corresponding fault type is obtained.

[0074] In this embodiment, by using a large model to determine the fault type, the efficiency of determining the fault type can be improved, and the subsequent steps (signaling analysis) can be facilitated. The large model used for signaling analysis can be used to perform analysis based on the determined fault type, thereby improving the accuracy of user report processing.

[0075] In one embodiment, there may be multiple processing flows obtained by "determining the processing flow corresponding to the user's declaration" in the above embodiment. For this, the signaling can be obtained and analyzed in sequence according to the sequence of the flows. For example, in the first processing flow, the corresponding signaling file is obtained and analyzed line by line to see if there is a problem with the current signaling file. If there is no problem signaling, the second processing flow is entered to obtain the corresponding signaling file and analyze line by line to see if there is a problem with the current signaling file. If there is no problem, the next processing flow is entered until all the signaling files corresponding to the processing flow corresponding to the user's declaration are analyzed, or the problem signaling has been demarcated and located.

[0076] In one embodiment, the user declaration processing method in the aforementioned embodiment also includes: when the fault type is a preset fault type, querying the setting information of the user terminal corresponding to the user declaration information, and checking whether the setting information of the user terminal is passed; the "calling the corresponding interface to obtain the signaling information corresponding to the fault type" in the aforementioned embodiment includes: when the fault type is a preset fault type, and the inspection result of the setting information of the user terminal is passed, calling the corresponding interface to obtain the signaling information corresponding to the fault type.

[0077] Exemplarily, the preset fault type may be, but is not limited to, a call fault, a network fault, and the like. The user terminal may be, but is not limited to, a mobile digital phone, a satellite phone, a smart communication tablet, a smart communication watch, and the like used by the user; the setting information of the user terminal is information related to the terminal settings used by the user, for example, turning on the high-definition call function of the terminal. Whether the setting information of the user terminal is passed, that is, whether the setting of the user terminal meets the requirements, for example, when the fault type is a call fault, it is necessary to determine the setting information of the corresponding user terminal, for example, to determine whether there is a 4G or 5G signal on the mobile phone, if not, the setting information of the user terminal is not passed, and it is necessary to check the ICCID (Integrated Circuit Card Identifier) ​​and replace the terminal; if yes, it is determined whether the high-definition call is turned on on the mobile phone. If the corresponding setting information indicates that the user terminal has turned on the high-definition call function of the terminal, the setting information of the user terminal is passed, otherwise, it is not passed.

[0078] In the case where the fault type is a preset fault type, query the setting information of the user terminal corresponding to the user's declaration information and check whether the setting information passes, that is, check whether there is a problem with the setting information. If the check fails, it means whether there is a problem with the setting information. Then the fault type corresponding to the user's declaration may be caused by incorrect settings of the user terminal (of course, it may also be caused by incorrect settings and signaling-related problems). Therefore, the setting information of the user terminal can be adjusted first without performing signaling analysis (of course, the signaling can also be analyzed to verify whether the check of the setting information is correct and whether there are problems with the signaling based on the results of the signaling analysis). If the check passes, the next step is to call the corresponding interface to obtain the signaling information corresponding to the fault type.

[0079] In this example, the user terminal setting information is queried and checked, and if the check passes, the corresponding signaling information is obtained. This adds a pre-processing, that is, first eliminating the fault caused by the user terminal setting error, and then obtaining the signaling and analyzing it. This can make the corresponding fault detection more comprehensive and improve the efficiency and accuracy of user report processing.

[0080] In one embodiment, after "obtaining the problem signaling corresponding to the user declaration information and the problems existing in the problem signaling" in the aforementioned embodiment, the user declaration processing method also includes: according to the problem signaling and the problems existing in the problem signaling, combined with the handling rule information in the preset knowledge base, obtaining corresponding handling suggestion information, and sending the handling suggestion information to the corresponding entity.

[0081] Exemplarily, the preset knowledge base also stores handling rule information, which is the corresponding handling method given for various possible problem signaling and problems existing in the problem signaling in the user's report. For example, for call failure, inability to make calls, call failures on all numbers, and unavailability in a small area, the corresponding handling rule information can be sent to wireless professionals for processing. From the handling rule information in the preset knowledge base, handling suggestion information corresponding to the problem signaling and the problems existing in the problem signaling can be obtained, and the handling suggestion information can be sent to the corresponding entity.

[0082] In this example, the obtained problem signaling and the problems existing in the problem signaling are combined with the handling rule information in the preset knowledge base to obtain corresponding handling suggestion information and send it to the corresponding entity. This enables the user report processing method to not only obtain the problem signaling and the problems existing in the problem signaling, but also automatically give corresponding handling suggestions, thereby further improving the processing efficiency and accuracy of user reports and avoiding possible errors and inefficiencies in manual processing.

[0083] In one embodiment, see Figure 3 As shown, a method for processing user declaration is provided. The method is applicable on the premise that the fault type corresponding to the user declaration information is a call fault, and includes steps S301 to S305. Figure 4 , Figure 4 The principle of the method is schematically explained. The details are as follows:

[0084] Step S301: Plan the process by combining the user declaration information and the business process in the local knowledge base. This step corresponds to "before semantic recognition of the user declaration information, determine the processing flow corresponding to the user declaration based on the user declaration information and the business process information in the preset knowledge base; determine at least one corresponding data interface based on the processing flow; call the data interface to obtain the signaling information corresponding to the fault type".

[0085] The business process here can be found in Figure 2 The content shown in the figure, the Di ability refers to the ability to identify whether the number status is abnormal, vMAX stands for Value Multi Analysis eXpert (value multi-dimensional analysis expert), and AS stands for Multimedia Telephony Application Server (multimedia telephony application server).

[0086] Step S302: semantic recognition. This step corresponds to "performing semantic recognition on the user's reported information to determine the type of fault reported by the user".

[0087] Based on the excellent understanding ability of the large model, the faults encountered by users are intelligently identified. For example, if a user reports that his call is automatically disconnected after about one minute, the semantic identification is a call fault. This step can be found in the previous related section and will not be repeated here.

[0088] Step S303: Query user settings. This step corresponds to "query the setting information of the user terminal corresponding to the user declaration information, and check whether the setting information of the user terminal is passed; if the check result of the setting information of the user terminal is passed, call the corresponding interface to obtain the signaling information corresponding to the fault type".

[0089] Querying user settings can be achieved through the terminal settings query interface. For example, in case of call failure, you need to first check whether the cause of the call failure is the user's mobile phone settings or a network problem. This step corresponds to "querying the setting information of the user terminal corresponding to the user's declaration information, and checking whether the setting information of the user terminal is passed; if the check result of the setting information of the user terminal is passed, call the corresponding interface to obtain the signaling information corresponding to the fault type."

[0090] It should be pointed out that Figure 4 When the content corresponding to the ellipsis indicates other types of faults, other processing flows other than step S303 may be corresponded. For example, when the fault type reported by the user is that the call cannot be made, the processing flow corresponding to step S303 may be to determine whether all numbers have call failures. If so, determine whether the call cannot be used in a small area or in most places. If the call cannot be used in a small area, it may be assigned to wireless professionals (terminals or personnel, etc.) for processing. If all numbers have call failures, the next step of signaling analysis and delimitation is executed.

[0091] Step S304: Signaling analysis and delimitation. This step corresponds to "calling the corresponding interface to obtain the signaling information corresponding to the fault type; inputting the signaling information into the preset signaling analysis model, and using the prompt engineering to guide the signaling analysis model to delimit and locate the signaling information according to the user's declaration information; based on the output of the signaling analysis model, obtaining the problem signaling corresponding to the user's declaration information and the problem existing in the problem signaling" and the relevant technical features in the corresponding dependent claims with a reference relationship.

[0092] The big model here can be a signaling analysis big model, which can be found in the above-mentioned related parts and will not be described here. When it is determined that there is no problem with the user settings, the signaling analysis and demarcation will begin. For example, if the fault type is a network problem, the corresponding interface will be called to obtain the signaling file and use the signaling analysis big model. When using this big model, it is necessary to use the promp project, input the rfc (Request For Comments) rules and specific signaling, and the Agent will intervene to locate the problem of the signaling. Regarding the prompt project, it can be based on the BROKE prompt word framework, as mentioned above, the framework can include five frames, and at the same time, there are three ways of "filling, selecting, and growing". "Filling": filling the missing broke content by interacting with the outside; "selecting": storing multiple templates in the local knowledge base for selection during actual use. The templates can be diverse, including templates that can meet the five frames at the same time, and templates that only meet the one, two, three, etc. frames; "growing": growth, dynamically generating broke content based on planning capabilities, and prompt combined with professional knowledge, for example, according to the user's declaration information and the preset knowledge base, the user's declaration processing flow can be planned in advance, and the corresponding prompt content can be dynamically generated according to the process capabilities. For example, the next process is signaling analysis, and the prompt content is automatically generated, "Now you have received a task, you need to perform signaling analysis, you are a professional signaling analyst, you need to analyze whether there is a problem with the current signaling". Please refer to the relevant parts of the aforementioned embodiments here. The following is a simple example. After analysis, the large model locates a problem in a line of signaling in the current signaling file, and analyzes that the specific problem of the signaling is that it loops between volte and MGCF, violating relevant rules, thereby completing the signaling analysis and delimitation.

[0093] Step S305: Next step is to give treatment suggestions based on the rules in the knowledge base. This step corresponds to "according to the problem signaling and the problems existing in the problem signaling, combined with the treatment rule information in the preset knowledge base, obtain the corresponding treatment suggestion information, and send the treatment suggestion information to the corresponding entity".

[0094] After signaling analysis and demarcation, the corresponding results can be obtained. Based on the results and combined with the relevant disposal rules in the knowledge base, disposal suggestions are given, for example, directly dispatching the order to the wireless side for disposal.

[0095] In the above user declaration processing method, the corresponding fault type is obtained by semantically identifying the user declaration information; the corresponding signaling information is analyzed by the signaling analysis model to determine the problem signaling, thereby realizing the automation of user declaration processing. In addition, by guiding the signaling analysis model through the prompt engineering and incorporating the BROKE prompt word framework into the prompt engineering, the output of the model is more in line with the needs of user declaration processing, and the problem signaling and the problems existing in the problem signaling are more accurate. Through the three methods of "filling, selecting, and length", the use of the prompt engineering to guide the signaling analysis model is diversified, thereby making the located problem signaling more accurate; at the same time, due to this diversity, there are multiple ways to use the prompt engineering to guide the signaling analysis model. When one method fails or the effect is not good (such as due to various unexpected situations), the remaining methods can be selected to realize the guidance of the prompt engineering to the signaling analysis model, thereby increasing the fault tolerance of the declaration processing method and improving the stability. By planning the process, avoiding the execution of unnecessary processes and obtaining unnecessary signaling information, the required signaling information corresponding to the fault type can be quickly and accurately obtained, thereby helping to improve the efficiency and accuracy of user declaration processing. By querying the user settings, we can first eliminate the faults caused by the wrong user terminal settings, and then obtain the signaling and analyze it, so that we can more comprehensively troubleshoot the corresponding faults and improve the efficiency and accuracy of user report processing. In addition to being able to obtain the problem signaling and the problems with the problem signaling, we can also automatically give corresponding disposal suggestions, thereby further improving the efficiency and accuracy of user report processing and avoiding the mistakes and inefficiencies that may occur in manual processing.

[0096] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0097] Based on the same inventive concept, the embodiment of the present application also provides a user declaration processing device for implementing the user declaration processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more user declaration processing device embodiments provided below can refer to the limitations of the user declaration processing method above, and will not be repeated here.

[0098] In an exemplary embodiment, Figure 6 As shown, a user declaration processing device 600 is provided, comprising:

[0099] Identification module 601, used to determine the fault type of the user's reported information;

[0100] An acquisition module 602 is used to call a corresponding interface to acquire signaling information corresponding to the fault type;

[0101] Analysis module 603, used for inputting signaling information into a preset signaling analysis model, and using the prompt engineering to guide the signaling analysis model to delimit and locate the signaling information according to the user declaration information;

[0102] The determination module 604 is used to obtain the problem signaling corresponding to the user's declaration information and the problems existing in the problem signaling based on the output of the signaling analysis large model.

[0103] In one of the embodiments, the analysis module 603 is also used for: the prompt project includes a BROKE prompt word framework; the prompt project is used to guide the signaling analysis large model, including at least one of the following: if the initial prompt content under the BROKE prompt word framework has missing content, the missing prompt content is filled in by interacting with the outside to obtain the target prompt content and input it into the signaling analysis large model; the signaling analysis prompt template corresponding to the fault type is determined from multiple preset prompt project templates, and the signaling analysis prompt template is input into the signaling analysis large model; multiple prompt project templates are set based on the BROKE prompt word framework; according to the user declaration information and the preset knowledge base, the signaling analysis prompt content for the fault type is generated based on the BROKE prompt word framework, and is input into the signaling analysis large model as the target prompt content.

[0104] In one of the embodiments, the analysis module 603 is further configured to determine a signaling analysis prompt template corresponding to the fault type from a plurality of preset prompt engineering templates based on a signaling analysis rule corresponding to the fault type.

[0105] In one of the embodiments, the analysis module 603 is further used to input the signaling analysis rule into the large model when it is determined that the signaling analysis prompt template does not contain the content of the signaling analysis rule.

[0106] In one of the embodiments, the identification module 601 is also used to determine the processing flow corresponding to the user report based on the user report information and the business process information in a preset knowledge base before determining the fault type of the user report information; the acquisition module 602 is also used to determine at least one corresponding data interface based on the processing flow; and call the data interface to obtain signaling information corresponding to the fault type.

[0107] In one of the embodiments, the identification module 601 is further used to input the user report information into a preset big model, and identify the fault type of the user report information through the semantics of the big model.

[0108] In one embodiment, the analysis module 603 is further configured to:

[0109] When the fault type is a preset fault type, query the setting information of the user terminal corresponding to the user declaration information, and check whether the setting information of the user terminal is passed; when the fault type is a preset fault type, if the inspection result of the setting information of the user terminal is passed, call the corresponding interface to obtain the signaling information corresponding to the fault type.

[0110] In one of the embodiments, the determination module 604 is also used to, after obtaining the problem signaling corresponding to the user's declaration information and the problems existing in the problem signaling, obtain corresponding handling suggestion information based on the problem signaling and the problems existing in the problem signaling, combined with the handling rule information in the preset knowledge base, and send the handling suggestion information to the corresponding entity.

[0111] Each module in the above user declaration processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0112] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store relevant data required for processing user declarations, such as a preset knowledge base, a preset plurality of prompt engineering templates and other data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a user declaration processing method is implemented.

[0113] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0114] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, any step in any of the above-mentioned embodiments is implemented.

[0115] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any step in any of the aforementioned embodiments is implemented.

[0116] In one embodiment, when the computer program is executed by a processor, any step in any of the foregoing embodiments is also implemented.

[0117] It should be noted that the user information (including but not limited to user device information, user personal information, user terminal configuration information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0118] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0119] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0120] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for processing user claims, characterized in that: The method comprises: Determine the fault type of the user's reported information; Call the corresponding interface to obtain the signaling information corresponding to the fault type; Input the signaling information into a preset large model, and use a prompting project to guide the large model to delimit and locate the signaling information according to the user declaration information; Based on the output of the large model, the problem signaling corresponding to the user declaration information and the problems existing in the problem signaling are obtained.

2. The method according to claim 1, characterized in that The prompt engineering includes a BROKE prompt word framework; the use of the prompt engineering to guide the large model includes at least one of the following: If the initial prompt content under the BROKE prompt word framework has missing content, the missing prompt content is filled in by interacting with the outside to obtain the target prompt content and input it into the large model; Determine a signaling analysis prompt template corresponding to the fault type from a plurality of preset prompt engineering templates, and input the signaling analysis prompt template into the large model; the plurality of prompt engineering templates are set based on a BROKE prompt word framework; According to the user declaration information and the preset knowledge base, signaling analysis prompt content for the fault type is generated based on the BROKE prompt word framework and input into the large model as the target prompt content.

3. The method according to claim 2, characterized in that The step of determining a signaling analysis prompt template corresponding to the fault type from a plurality of preset prompt engineering templates includes: Based on the signaling analysis rule corresponding to the fault type, a signaling analysis prompt template corresponding to the fault type is determined from a plurality of preset prompt engineering templates.

4. The method according to claim 3, characterized in that: The method further comprises: When it is determined that the signaling analysis prompt template does not contain the content of the signaling analysis rule, the signaling analysis rule is input into the large model.

5. The method according to claim 1, characterized in that Before determining the fault type of the user's reported information, the method further includes: Determine the processing flow corresponding to the user's declaration based on the user's declaration information and the business process information in the preset knowledge base; The calling of the corresponding interface to obtain signaling information corresponding to the fault type includes: Determining at least one corresponding data interface based on the processing flow; The data interface is called to obtain signaling information corresponding to the fault type.

6. The method according to claim 1, characterized in that The determining of the fault type of the user's reported information includes: The user declaration information is input into a preset big model, and the fault type of the user declaration information is identified through the semantics of the big model.

7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: In the case where the fault type is a preset fault type, querying the setting information of the user terminal corresponding to the user declaration information, and checking whether the setting information of the user terminal is passed; The calling of the corresponding interface to obtain signaling information corresponding to the fault type includes: In the case where the fault type is a preset fault type and the check result of the setting information of the user terminal is passed, a corresponding interface is called to obtain signaling information corresponding to the fault type.

8. The method according to any one of claims 1 to 6, characterized in that: After obtaining the problem signaling corresponding to the user declaration information and the problem existing in the problem signaling, the method further includes: According to the problem signaling and the problem existing in the problem signaling, combined with the processing rule information in the preset knowledge base, corresponding processing suggestion information is obtained, and the processing suggestion information is sent to the corresponding entity.

9. A user claim processing device, characterized in that: The device comprises: An identification module, used to determine the fault type of the user's reported information; An acquisition module, used to call a corresponding interface to obtain signaling information corresponding to the fault type; An analysis module, used for inputting the signaling information into a preset large model, and using a prompting project to guide the large model to delimit and locate the signaling information according to the user declaration information; A determination module is used to obtain the problem signaling corresponding to the user declaration information and the problems existing in the problem signaling based on the output of the large model.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images