Application analysis processing method and device

Abstract

The embodiment of the invention provides an application analysis processing method and device.The application analysis processing method comprises the steps that in the application analysis processing process, after analysis elements for conducting application analysis on an application to be analyzed are obtained, on one hand, the analysis elements and problem generation cues are input into a large language model for problem generation, and on the other hand, the analysis elements and the problem generation cues are input into the large language model for problem generation; on the other hand, querying target interaction data in the user interaction data of the application to be analyzed according to the analysis elements; and inputting the target interaction data, the application analysis cue word and at least one application analysis problem in the application analysis list into a large language model for application analysis to obtain an application analysis result, thereby realizing application analysis of the to-be-analyzed application.

Description

TECHNICAL FIELD

[0001] The present document relates to the technical field of artificial intelligence, and in particular to an application analysis processing method and device. BACKGROUND

[0002] With the emergence of more and more applications, the competitiveness of the applications is the focus of attention of institutions or application providers; collecting relevant data of users through offline methods to understand the information such as the demand, preference and behavior pattern of users for the applications can help the institutions or application providers to better understand the application demand and formulate application strategies to improve the value and competitiveness of the applications; in this process, systematic data collection can be performed through methods such as questionnaire survey, interview and observation, and then the collected data is analyzed to understand the information such as the demand, preference and behavior pattern of users for the applications. SUMMARY

[0003] One or more embodiments of the present specification provide an application analysis processing method, comprising: acquiring an analysis element for application analysis of a to-be-analyzed application. Inputting the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis question list. Querying target interaction data in user interaction data of the to-be-analyzed application according to the analysis element. Inputting the target interaction data, an application analysis prompt word and at least one application analysis question in the application analysis question list into the large language model for application analysis to obtain an application analysis result.

[0004] One or more embodiments of the present specification provide an application analysis processing device, comprising: an analysis element acquisition module configured to acquire an analysis element for application analysis of a to-be-analyzed application. A question generation module configured to input the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis question list. A data query module configured to query target interaction data in user interaction data of the to-be-analyzed application according to the analysis element. An application analysis module configured to input the target interaction data, an application analysis prompt word and at least one application analysis question in the application analysis question list into the large language model for application analysis to obtain an application analysis result.

[0005] The one or more embodiments of the specification provide an application analysis processing device, comprising: a processor; and a memory configured to store computer executable instructions which, when executed, cause the processor to: acquire an analysis element for application analysis of an application to be analyzed. Input the analysis element and question generation prompt words into a large language model for question generation to obtain an application analysis question list. Query target interaction data in user interaction data of the application to be analyzed according to the analysis element. Input the target interaction data, application analysis prompt words, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result.

[0006] The one or more embodiments of the specification provide a computer readable storage medium for storing computer executable instructions, which, when executed, implement the following processes: acquiring an analysis element for application analysis of an application to be analyzed. Input the analysis element and question generation prompt words into a large language model for question generation to obtain an application analysis question list. Query target interaction data in user interaction data of the application to be analyzed according to the analysis element. Input the target interaction data, application analysis prompt words, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result. BRIEF DESCRIPTION OF DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the one or more embodiments of the specification or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the specification, and those skilled in the art can also obtain other drawings according to these drawings without creative labor; Figure 1 A schematic diagram of an application analysis processing method implementation environment provided by the one or more embodiments of the specification; Figure 2 A flowchart of an application analysis processing method processing process provided by the one or more embodiments of the specification; Figure 3 A flowchart of an application analysis processing method processing process provided by the one or more embodiments of the specification for a product guarantee analysis scenario; Figure 4 A schematic diagram of an application analysis processing device embodiment provided by the one or more embodiments of the specification; Figure 5 A structural schematic diagram of an application analysis processing device provided by the one or more embodiments of the specification. DETAILED DESCRIPTION

[0008] In order to make the person skilled in the art better understand the technical solutions in one or more embodiments of the present specification, the technical solutions in one or more embodiments of the present specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of the present specification. Obviously, the described embodiments are only a part of the embodiments of the present specification, rather than all the embodiments. Based on one or more embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present document.

[0009] The application analysis processing method provided by one or more embodiments of the present specification can be applied to the implementation environment of the application analysis system, which is described with reference to Figure 1 The implementation environment at least includes: a server 101 and a large language model 102. Among them, the server 101 is used to generate an application analysis question list by using the large language model 102 according to the analysis elements in the application analysis processing process, query target interaction data in the user interaction data of the application to be analyzed according to the analysis elements, and perform application analysis by using the large language model 102 according to the application analysis question list, target interaction data and application analysis prompt words to obtain the application analysis result; the server 101 can be a server, or a server cluster composed of several servers, or one or more cloud servers in a cloud computing platform; The large language model 102 is used for question generation and application analysis; the large language model 102 can be a large language model (LLM, Large Language Models); In addition, the implementation environment can also include a user terminal 103 and a database 104. Among them, the user terminal 103 includes the user terminal of the analysis user who submits the analysis elements, and is used for submitting the analysis elements and displaying the application analysis result; the user terminal 103 can be a mobile phone, a personal computer, a tablet computer, an electronic book reader, a device for information interaction based on VR (Virtual Reality, Virtual Reality Technology), a vehicle terminal, an IoT device, a wearable smart device, a laptop computer and a desktop computer, etc. The database 104 is used to store the user interaction data of the application to be analyzed.

[0010] In the implementation environment, after the server 101 obtains the analysis elements submitted by the user terminal 103 for application analysis of the application to be analyzed, the server 101 first inputs the analysis elements and the question generation prompt word into the large language model 102 for question generation to obtain a list of application analysis questions, and then queries target interaction data in the database 104 storing the user interaction data of the application to be analyzed according to the analysis elements, and then inputs the target interaction data, the application analysis prompt word, and at least one application analysis question in the list of application analysis questions into the large language model 102 for application analysis according to the application analysis question to obtain an application analysis result. In this way, the application analysis of the application to be analyzed is implemented.

[0011] One or more implementations of the application analysis processing method provided in the specification are as follows: With reference to Figure 2 The application analysis processing method provided in the embodiment specifically includes steps S202 to S208.

[0012] In step S202, analysis elements for application analysis of an application to be analyzed are obtained.

[0013] The application in the embodiment includes an application that can be used by a user to participate in or interact with the user, such as an application program, a subprogram, a service, a project, and / or a product. The application analysis in the embodiment includes data statistics and / or data recognition of the application in at least one dimension to obtain relevant information such as evaluation and / or demand of the application, such as application research. The application analysis in the embodiment can be application evaluation analysis, or application perception analysis, or application demand analysis, or application analysis including application evaluation analysis, application perception analysis, and / or application demand analysis, which is not limited in the embodiment. In addition to application evaluation analysis, application perception analysis, and application demand analysis, the application analysis can be in other dimensions, which is not limited in the embodiment. The application to be analyzed in the embodiment includes an application to be analyzed, such as a security project or a security product participating in research.

[0014] It should be noted that the application in the embodiment can be replaced by a program, a service, or a product, and the other features related to the application can also be replaced by features related to the program, the service, or the product. For example, the application to be analyzed is replaced by a program to be analyzed, a service to be analyzed, or a product to be analyzed, and the application analysis is replaced by program analysis, service analysis, or product analysis. In addition, the application in the embodiment can be replaced by an object, and the other features related to the application can also be replaced by features related to the object. For example, the application to be analyzed is replaced by an object to be analyzed, and the application analysis is replaced by object analysis.

[0015] It should be noted that the analysis in the embodiment can also be replaced by research or data statistics, and the other features related to the analysis can also be replaced by the features related to the data statistics research, such as application research, application research questions, or product research, product research questions.

[0016] The analysis element in the embodiment refers to key information describing application analysis, such as analysis purpose of performing application analysis, analysis object requirement of participating in application analysis, and / or analysis content of performing application analysis; optionally, the analysis element includes at least one of the following: analysis category, analysis object feature, and question element. The analysis category is used to represent the analysis purpose, the analysis object feature is used to represent the analysis object requirement of participating in the application analysis, and the question element is used to represent the analysis content of performing the application analysis.

[0017] In specific implementation, the analysis element of performing application analysis on the application to be analyzed is obtained. For example, the analysis element of performing product analysis on the guarantee product is obtained; wherein the analysis element includes: analysis category: user satisfaction with the guarantee product; analysis object feature: user who purchased the guarantee product in the past 3 months to 6 months; question element: what is the reason for purchasing the guarantee product, what is the content that is more satisfied after purchasing the guarantee product, and what is the content that is not satisfied.

[0018] For another example, the analysis element of performing program analysis on the guarantee application program is obtained; wherein the analysis element includes: analysis category: user understanding of the guarantee application program; analysis object feature: 18-60 year-old users in first-tier and second-tier cities; question element: whether the user understands the guarantee product in the guarantee application program, and what is the channel based on which the user understands the guarantee product in the guarantee application program.

[0019] In step S204, the analysis element and the question generation prompt word are input into the large language model to generate questions, and a list of application analysis questions is obtained.

[0020] In specific implementation, after obtaining the analysis element of performing application analysis on the application to be analyzed, in order to improve the effectiveness and comprehensiveness of the application analysis, the analysis element and the question generation prompt word are input into the large language model to generate questions, and a list of application analysis questions is obtained, so as to perform application analysis according to the application analysis questions in the list of application analysis questions.

[0021] The application analysis question in the embodiment includes a question designed for application analysis with the application to be analyzed as the question subject; includes a question for data statistics or data summary with the application to be analyzed as the analysis subject, so as to perform application analysis on the application to be analyzed; for example, "what does the user purchase the application to be analyzed for?", and "how is the user satisfaction distribution of the application to be analyzed?".

[0022] It should be noted that the application analysis question can be replaced by a statistical analysis question or a summary analysis question, such as a statistical analysis question for data statistics of an application to be analyzed or a summary analysis question for summary analysis. The list composed of one or more application analysis questions is an application analysis question list.

[0023] The question generation prompt word is used to instruct the large language model to generate a question for statistical analysis or summary analysis of the application to be analyzed according to the analysis element. In order to improve the effectiveness and accuracy of the application analysis question generated by the large language model, the application introduction data of the application to be analyzed can also be included in the question generation prompt word. Specifically, the application introduction data of the application to be analyzed can be written into a question generation prompt template to obtain the question generation prompt word.

[0024] In the specific implementation process, in order to improve the accuracy of the generated application analysis question, the large language model can first generate at least one initial question as an application analysis question according to the question element in the analysis element. In order to improve the effectiveness and comprehensiveness of the application analysis question, on the basis of obtaining at least one initial question, the question can be extended according to the initial question and the analysis category in the analysis element to obtain at least one associated question together with the initial question as an application analysis question, and then an application analysis question list is obtained; in an optional implementation manner provided by the embodiment, the following operations can be performed in the process of question generation: read the question element in the analysis element according to the question generation prompt word, and generate at least one initial question according to the question element; extend the question according to the at least one initial question and the analysis category in the analysis element to obtain at least one associated question; generate an application analysis question list according to the at least one initial question and the at least one associated question.

[0025] Specifically, after the large language model obtains the analysis element and the question generation prompt word, it first reads the question element in the analysis element according to the question generation prompt word, and then generates at least one initial question according to the question element. At this point, the application analysis question can be generated according to the at least one initial question. Further, the question can be extended according to the initial question and the analysis category in the analysis element to obtain at least one associated question, and then an application analysis question list is generated according to the at least one initial question and the at least one associated question.

[0026] Taking the analysis elements for product analysis of the guarantee product as an example, the problem elements in the analysis elements are: what is the reason for purchasing the product, what is the content that the user is more satisfied with after purchasing the product, and what is the content that the user is not satisfied with; after the large language model obtains the analysis elements and the problem generation prompt words, first, according to the problem elements, the initial questions for application analysis questioning are generated: “what does the user purchase the guarantee product for?”, “what is the content that the user is more satisfied with after purchasing the guarantee product, and what is the content that the user is not satisfied with?”; then, the problems are extended from the initial questions and the analysis category of the user's satisfaction degree for the guarantee product to obtain the associated questions for application analysis questioning: “distribution of the user's purchase reason for the guarantee product”, “distribution of the user's satisfaction degree for the guarantee product”; then, the application analysis question list including the initial questions and the associated questions is generated.

[0027] In the process of generating the application analysis question list according to the at least one initial question and the at least one associated question, in addition to the above-mentioned generation of the application analysis question list by taking the initial questions and the associated questions as the application analysis questions, the initial questions and the associated questions can also be processed by merging to obtain the application analysis questions, and then the application analysis question list composed of the application analysis questions is obtained; in the specific execution process, the initial questions and the associated questions can be processed by merging according to the question correlation.

[0028] Specifically, after the initial questions and the associated questions are obtained, the initial questions and the associated questions are combined to obtain a question group (one question group contains one initial question and one associated question), the question correlation scores of the initial questions and the associated questions in each question group are calculated respectively, if the question correlation score is greater than the correlation score threshold, the initial question and the associated question in the corresponding question group are processed by merging to obtain the application analysis question, and if the question correlation score is equal to or less than the correlation score threshold, the initial question and the associated question in the corresponding question group are taken as the application analysis questions respectively.

[0029] With the above example, after obtaining the initial question and the associated question, the initial question and the associated question are combined to obtain 4 question groups, and the question relevance scores of the initial question and the associated question in each question group are calculated. After calculation, the question relevance score of the initial question and the associated question in the question group of “What are the reasons for the user to purchase the protection product? What is the distribution of the reasons for the user to purchase the protection product?” is greater than the relevance score threshold, and then the initial question and the associated question in the question group are merged to obtain the application analysis question “What are the reasons for the user to purchase the protection product? What is the distribution of the reasons for the user to purchase the protection product?”; Since the question relevance scores of the other question groups are less than or equal to the relevance score threshold, the initial question and the associated question in the other question groups are respectively taken as the application analysis question to obtain the following application analysis question list: 1、What are the reasons for the user to purchase the protection product? What is the distribution of the reasons for the user to purchase the protection product? 2、What are the contents that the user is more satisfied with after purchasing the protection product? What are the contents that the user is not satisfied with? 3、What is the satisfaction distribution of the user to the protection product.

[0030] It should be noted that the application analysis question list in this embodiment can also be an application analysis questionnaire for application analysis questioning; that is, the application analysis question list can be replaced by an application analysis questionnaire, and correspondingly, the features related to the application analysis question list in this embodiment can also be replaced by the features related to the application analysis questionnaire.

[0031] The large language model in this embodiment includes an LLM (Large Language Model). In addition, the large language model in this embodiment can also be a natural language model pre-trained, and the large language model can adopt a foundation model or a pre-trained model, and the specific architecture of the large language model can be a neural network architecture with a large number of parameters, a Transform architecture, or other architectures. In the specific execution process, the large language model can directly adopt a foundation model or a pre-trained model, and can also fine-tune (Supervised Fine-Tuning, SFT) the foundation model or the pre-trained model for a question generation task and / or an application analysis task based on the foundation model or the pre-trained model, that is, a large language model capable of question generation and / or application analysis can be obtained.

[0032] In a specific implementation, after obtaining the application analysis question list, in order to improve the effectiveness and accuracy of the application analysis questions in the application analysis question list, the analysis user who submits the analysis elements can be sent the application analysis question list for confirmation of the application analysis question list; the analysis user can also edit and confirm the application analysis questions in the application analysis question list; in an optional implementation provided by the embodiment, after obtaining the application analysis question list, the following operations can also be performed: sending the application analysis question list to the analysis user; obtaining a revised application analysis question list according to a revised confirmation instruction submitted by the analysis user after editing the application analysis question list.

[0033] Specifically, after inputting the analysis elements and the question generation prompt words into the large language model to generate questions and obtaining the application analysis question list, the analysis user can be sent the application analysis question list; if a confirmation instruction of the analysis user on the application analysis question list is detected, the following steps S206 to S208 are performed; if a revised confirmation instruction submitted by the analysis user after editing the application analysis question list is detected, a revised application analysis question list is obtained.

[0034] The above provides a specific process of question generation by a large language model, in addition, question generation can also be performed without being based on a large language model; that is, step S204 can be replaced by generating questions according to the analysis elements to obtain an application analysis question list, and the application analysis question list and other one or more processing steps provided by the embodiment form a new implementation. Optionally, in the process of generating questions according to the analysis elements, the questions can be first converted according to the question elements in the analysis elements to obtain at least one initial question, then the at least one initial question and the analysis category in the analysis elements are used to extend the questions to obtain at least one associated question, and finally the at least one initial question and the at least one associated question are used to generate the application analysis question list.

[0035] Step S206: querying target interaction data in the user interaction data of the application to be analyzed according to the analysis elements.

[0036] In a specific implementation, after obtaining the analysis elements for application analysis of the application to be analyzed, on the one hand, the analysis elements and the question generation prompt words are input into the large language model to generate questions and obtain an application analysis question list for application analysis questioning, and on the other hand, target interaction data for answering the application analysis questions is queried in the user interaction data of the application to be analyzed according to the analysis elements.

[0037] In the specific implementation process, the target interaction data matching the analysis object feature in the analysis element can be queried and analyzed in the user interaction data of the application to be analyzed, so that the target interaction data for answering the application analysis question is obtained from the analysis element, and the accuracy of the determined target interaction data and the matching degree with the analysis element are improved. In an optional implementation provided by the embodiment, the process of querying the target interaction data in the user interaction data of the application to be analyzed according to the analysis element is implemented in the following manner: reading the analysis object feature in the analysis element; querying the target interaction data matching the analysis object feature in the user interaction data.

[0038] Optionally, the analysis object feature includes a time feature and / or a user reference feature.

[0039] The user reference feature can include a user geographic area, user resource data, a user growth period, a user category (male or female), an application identifier of a purchased application, and / or a purchase time.

[0040] Specifically, the target interaction data is queried in the user interaction data according to the analysis object feature in the analysis element.

[0041] For example, the analysis object feature in the analysis element is that a user who has purchased the guarantee product in the past 3 to 6 months, and the user interaction data of the user who has purchased the guarantee product in the past 3 to 6 months is queried as the target interaction data in the user interaction data of the guarantee product.

[0042] In the embodiment, the user interaction data includes user dialogue data and / or user reference data. The user dialogue data can be user dialogue data generated by a customer service or a service user (application sales user) of the application to be analyzed in a dialogue with a user. The user dialogue data can include dialogue content, dialogue scene, and / or dialogue role. The dialogue role can include an intelligent customer service, a manual customer service, and / or a service user. The user reference data can be related data of the user, such as a user geographic area, user resource data, a user growth period, a user category, an application identifier of a purchased application, and / or a purchase time.

[0043] In an optional implementation provided by the embodiment, the user interaction data can be obtained in the following manner: obtaining user dialogue data related to the application to be analyzed, and reading user reference data of a user associated with the user dialogue data; taking the user reference data and the user dialogue data as the user interaction data.

[0044] Specifically, first, user dialogue data related to the application to be analyzed can be obtained, then user reference data of the user associated with the user dialogue data can be read, and the user reference data can be used as a label of the user dialogue data, and the user dialogue data labeled with the user reference data can be obtained as the user interaction data. The user dialogue data can be dialogue voice generated by voice dialogue with the user, can be dialogue text generated by instant message dialogue with the user, can be user evaluation data of the user to the application to be analyzed, and can be user dialogue data obtained by dialogue with the user through other interaction channels. The embodiment is not limited in this regard.

[0045] It should be noted that in the present embodiment, the user interaction data of the application to be analyzed can be stored through a database (for example, a user interaction database). The database can store a plurality of user interaction data, and the user interaction data in the user interaction database can be obtained in the above manner. The embodiment is not limited in this regard.

[0046] In the present embodiment, step S204 and step S206 can be executed in the order of step S204 first and then step S206, or in the order of step S206 first and then step S204, or simultaneously. The present embodiment does not limit the execution order of step S204 and step S206.

[0047] In step S208, the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list are input into the large language model to perform application analysis, and an application analysis result is obtained.

[0048] In specific implementation, after obtaining the application analysis question list for application analysis and the target interaction data for answering the application analysis question, the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list are input into the large language model to perform application analysis, and an application analysis result is obtained. In this way, the large language model performs application analysis according to the application analysis question and the target interaction data, improves the accuracy and effectiveness of the obtained application analysis result, and improves the convenience of application analysis.

[0049] In the present embodiment, the large language model used for application analysis can be the same as the large language model used for question generation, or can be different. In the process of specifically performing application analysis, the question answers of the application analysis questions can be generated according to the answer generation mode corresponding to the question type of each application analysis question in the application analysis question list, and each application analysis question and the question answer are taken as the application analysis result; for example, for the application analysis question of the statistical analysis type, the target interaction data is classified according to the application analysis question of the statistical analysis type, and the application analysis result is generated according to the category interaction data under each statistical category; for example, for the application analysis question of the recognition type, the target interaction data is recognized according to the application analysis question of the recognition type, and the recognition result is taken as the application analysis result.

[0050] The following specifically describes the process of application analysis of the large language model for the application analysis question of the statistical analysis type and the recognition type.

[0051] (1) Statistical analysis type In specific implementation, in the process of application analysis of the large language model according to the application analysis question, the target interaction data and the application analysis prompt word, for the statistical analysis question, first, the target interaction data is classified according to the statistical category corresponding to the statistical analysis question, and the category interaction data under each statistical category is obtained, and then the application analysis result is generated according to the category interaction data under each statistical category.

[0052] In an optional implementation provided by the embodiment, in the process of application analysis of the large language model, the following operations can be performed: The statistical analysis question in the at least one application analysis question is read according to the application analysis prompt word; The target interaction data is classified based on the statistical analysis question, and the category interaction data under at least one statistical category is obtained; The application analysis result is generated according to the category interaction data under each statistical category.

[0053] Optionally, the statistical category includes the statistical category corresponding to the statistical analysis question. For example, the statistical category corresponding to the evaluation statistical question includes each evaluation category; the statistical category corresponding to the recommendation channel (perception channel) statistical question includes each recommendation channel.

[0054] Specifically, if the statistical analysis question is an evaluation statistical question, in the process of classifying the target interaction data according to the statistical analysis question to obtain the category interaction data under at least one category, the user evaluation data in the target interaction data can be first read according to the evaluation statistical question, and then the user evaluation data is classified according to the evaluation, and the user evaluation data under each evaluation category is obtained. Further, in the process of generating the application analysis result according to the category interaction data under each statistical category, the proportion of evaluation corresponding to each evaluation category can be calculated according to the total evaluation number and the user evaluation number under each evaluation category, and then the answer to the evaluation statistical question can be generated as the application analysis result according to the category identifier and the proportion of evaluation corresponding to each evaluation category.

[0055] For example, the application analysis question is the evaluation statistical question of "user satisfaction distribution of the guarantee product", and in the process of application analysis based on the evaluation statistical question, the large language model reads the user evaluation data in the target interaction data, then classifies the user evaluation data according to the evaluation category in the user evaluation data, obtains the user evaluation data under each evaluation category, then calculates the total evaluation number and the user evaluation number under each evaluation category according to the user evaluation data under each evaluation category, and then calculates the quotient of the user evaluation number under each evaluation category and the total evaluation number as the proportion of evaluation corresponding to each evaluation category. Finally, the satisfaction distribution is generated as the answer to the evaluation statistical question, that is, the application analysis result of the evaluation statistical question, according to the category identifier and the proportion of evaluation corresponding to each evaluation category.

[0056] In addition, if the statistical analysis question is a recommendation channel statistical question, in the process of classifying the target interaction data according to the statistical analysis question to obtain the category interaction data under at least one category, the recommendation channel data of each target interaction data can be read according to the recommendation channel statistical question, and the target interaction data can be classified according to the recommendation channel data to obtain the target interaction data under each recommendation channel. Further, in the process of generating the application analysis result according to the category interaction data under each statistical category, the proportion of evaluation corresponding to each evaluation category can be calculated according to the total evaluation number and the user evaluation number under each evaluation category, and then the answer to the evaluation statistical question can be generated as the application analysis result according to the category identifier and the proportion of evaluation corresponding to each evaluation category.

[0057] It should be noted that the obtained user evaluation data can be text evaluation data or voice evaluation data, and if it is voice evaluation data, voice recognition can be performed on the voice evaluation data to obtain text evaluation data. Similarly, the recommendation channel data can be text recommendation channel data or voice recommendation channel data, and if it is voice recommendation channel data, voice recognition can be performed on the voice recommendation channel data to obtain text recommendation channel data.

[0058] (2) Recognition type In particular implementation, in the process of application analysis by the large language model according to the application analysis problem, the target interaction data and the application analysis prompt word, for the identification problem, the target interaction data is identified according to the application analysis problem of the identification type, and the identification result is obtained as the application analysis result.

[0059] In an optional implementation provided by the embodiment, in the process of application analysis by the large language model, the following operations can be performed: read the identification problem in the application analysis problem according to the application analysis prompt word; read the target dialogue data in the target interaction data based on the identification problem, identify the target dialogue data, and obtain the identification result as the application analysis result.

[0060] Optionally, the target dialogue data includes user dialogue data carrying an identification keyword.

[0061] In the embodiment, the identification can be demand identification, reason identification, satisfactory content identification and / or unsatisfactory content identification; the above examples of identification are only illustrative, and other dimensions of identification can be performed according to actual needs, which are not limited in the embodiment.

[0062] Specifically, for the identification problem, first, the target dialogue data carrying the identification keyword in the target interaction data is read based on the identification problem, and then the target dialogue data is identified to obtain the identification result as the application analysis result.

[0063] For example, for the demand identification problem, the user dialogue data carrying the demand identification keyword in the target interaction data is read as the target dialogue data based on the demand identification problem, the demand identification is performed according to the target dialogue data, and the demand identification result is obtained as the answer to the demand identification problem, that is, as the application analysis result of the demand identification problem.

[0064] The application analysis of the application analysis problem of the statistical analysis type and the identification type is described above, in addition, the application analysis problem can also be of the identification statistical type, for the application analysis problem of the identification statistical type, identification can be performed first in the process of application analysis, and then statistical analysis is performed according to the identification result after the identification result is obtained, and the application analysis result is obtained. In an optional implementation provided by the embodiment, in the process of application analysis by the large language model, for an application analysis problem of the recognition statistical type, i.e., a recognition statistical problem, the large language model first reads target dialogue data carrying a recognition keyword in the target interaction data based on the recognition statistical problem, and then performs recognition on the target dialogue data to obtain a recognition result; then, the large language model classifies the recognition result according to at least one statistical category corresponding to the recognition statistical problem to obtain a category recognition result in at least one statistical category, and generates an application analysis result according to the category recognition result in each statistical category.

[0065] For example, for a demand recognition statistical problem, the large language model reads user dialogue data carrying a demand recognition keyword in the target interaction data as target dialogue data based on the demand recognition statistical problem, performs demand recognition according to the target dialogue data to obtain a demand recognition result, classifies the demand recognition result according to a demand category in the demand recognition result to obtain a demand recognition result corresponding to each demand category, calculates a total demand number and a demand number corresponding to each demand category according to the demand recognition result corresponding to each demand category, calculates a category proportion corresponding to each demand category according to the total demand number and the demand number corresponding to each demand category, and generates an answer to the demand recognition statistical problem as an application analysis result of the demand recognition statistical problem according to the category identifier and the category proportion corresponding to each demand category.

[0066] In addition, the application analysis problem can also include an application analysis problem of the statistical recognition type. For the application analysis problem of the statistical recognition type, the large language model can first perform statistical analysis to obtain a statistical analysis result, and then perform recognition on the statistical analysis result to obtain a recognition result as an answer to the statistical recognition problem, i.e., an application analysis result of the statistical recognition problem. The specific execution process can refer to the related content described above, and will not be described herein again.

[0067] In specific implementation, in order to further improve the effectiveness and accuracy of application analysis, in addition to inputting at least one application analysis problem in the application analysis problem list, the target interaction data, and the application analysis prompt word into the large language model, the analysis element can also be input into the large language model. Optionally, the application analysis prompt word contains the analysis element. The application analysis prompt word is obtained by writing the analysis element into the application analysis prompt template; In the case of inputting the target interaction data, the application analysis prompt word containing the analysis element, and at least one application analysis problem in the application analysis problem list into the large language model for application analysis, in order to improve the accuracy of application analysis and avoid the situation that the data number of the target interaction data queried in the user interaction data does not meet the data condition in the analysis element or does not match the analysis object feature in the analysis element, resulting in inaccurate application analysis, in an optional implementation provided by the embodiment, the large language model can perform application analysis in the following manner: The reading application analysis prompt word contains the analysis object feature in the analysis element, and the data condition is determined according to the analysis object feature; Detect whether the target interaction data meets the data condition; If not, data simulation is performed according to the target interaction data and the analysis object feature to obtain simulated interaction data; and the application analysis result of the application analysis question is obtained according to the simulated interaction data and the target interaction data.

[0068] Specifically, the analysis object feature in the analysis element contained in the application analysis prompt word is read, and the data condition is determined according to the analysis object feature. If the data distribution and / or the data number of the target interaction data meet the data condition, or if the target interaction data matches the analysis object feature, the application analysis of the application analysis question is performed according to the target interaction data, that is, the answer generation, to obtain the application analysis result. If the target interaction data does not meet the data condition, or if the target interaction data does not match the analysis object feature, data simulation is performed according to the target interaction data and the analysis object feature to obtain simulated interaction data, and then the application analysis of the application analysis question is performed according to the target interaction data and the simulated interaction data to obtain the application analysis result. The process of performing the application analysis of the application analysis question according to the target interaction data or performing the application analysis of the application analysis question according to the target interaction data and the simulated interaction data can be performed according to the application analysis process of the above different question types, which will not be described herein again.

[0069] For example, the analysis object feature is 1000 users who have purchased the guarantee product in the past 3 months to 6 months; the data condition determined according to the analysis object feature is: at least 1000 target interaction data, and the user corresponding to the target interaction data has purchased the guarantee product in the past 3 months to 6 months; then it is detected whether the data number of the target interaction data is greater than or equal to 1000, and whether the purchased guarantee product in the user reference data in each target interaction data contains the guarantee product and the purchase time is in the past 3 months to 6 months; after detection, the purchased guarantee product in the user reference data in each target interaction data contains the guarantee product, and the purchase time is in the past 3 months to 6 months, but the data number of the target interaction data is 800, then 200 simulated interaction data are obtained by performing data simulation according to the 800 target interaction data, and the application analysis of each application analysis question is performed according to the target interaction data and the simulated interaction data to obtain the application analysis result.

[0070] It should be noted that the application analysis in the embodiment can include answer generation according to the problem type of each application analysis question and the target interaction data and / or simulated interaction data, and obtaining the question answer of each application analysis question as the application analysis result. That is, step S208 can be replaced by: inputting the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list into the large language model to generate answers for each application analysis question, and obtaining the question answer of each application analysis question as the application analysis result, and forming a new implementation mode with other one or more processing steps provided in the embodiment.

[0071] Corresponding to the above-mentioned obtaining of the application analysis question list, the target interaction data, the application analysis prompt word, and at least one modified application analysis question in the modified application analysis question list can be input into the large language model to perform application analysis and obtain the application analysis result.

[0072] In specific implementation, in order to further improve the effectiveness of the analysis user performing application analysis, on the basis of generating answers for at least one application analysis question in the application analysis question list based on the target interaction data by the large language model, and obtaining the answer data of each application analysis question as the application analysis result, at least one interaction function can be further provided, so that the analysis user can perform interactive analysis through different interaction functions, and obtain more diverse, more comprehensive and more accurate analysis results.

[0073] In the specific execution process, after obtaining the application analysis result, the analysis user can be shown the application analysis result, and in the process of showing the application analysis result, the analysis user can be shown the interaction category, and the analysis user can select the interaction category to perform corresponding interactive analysis.

[0074] In an optional implementation provided in the embodiment, the interactive analysis data submitted by the analysis user after selecting the interaction category is obtained, and then the interactive analysis data and the interactive prompt word corresponding to the interaction category are input into the large language model to generate response data and obtain the response data. Optionally, the interaction category includes a question and answer category, a data generation category, and / or an analysis category. The data generation category can be replaced by a data simulation category. It should be noted that the response data can be displayed after being obtained, so as to improve the perception of the analysis user on the response data.

[0075] The interactive analysis processes under the question and answer category, the data generation category, and the analysis category are described below.

[0076] (1) Question and answer category In a specific implementation, if the question submitted after the analysis user selects the question and answer category is obtained, in an optional implementation provided by the embodiment, the question submitted after the analysis user selects the question and answer category and the answer generation prompt word corresponding to the question and answer category are input into the large language model to generate an answer, and the answer of the question is obtained.

[0077] In a specific implementation process, in order to improve the perception of the application analysis result, the question, the application analysis result and the answer generation prompt word are input into the large language model to generate an answer, and the answer of the question is obtained. In addition, the application analysis result can be written into the answer generation prompt template to obtain the answer generation prompt word, or the answer generation prompt word can be used to instruct the large language model to generate an answer of the question according to the previously generated application analysis result.

[0078] In addition, the target interaction data or the user interaction data of the application to be analyzed, the answer generation prompt word and the question submitted after the analysis user selects the question and answer category can be input into the large language model to generate an answer of the question according to the target interaction data or the user interaction data, and the answer of the question is obtained.

[0079] (2) Data generation category In a specific implementation, in order to improve the richness of the user interaction data and avoid too little target interaction data affecting the application analysis result, data generation can be performed to obtain simulated interaction data.

[0080] In a specific implementation process, if the generation element submitted after the analysis user selects the data generation category is obtained, in an optional implementation provided by the embodiment, first, the associated interaction data is queried in the user interaction data according to the generation element submitted after the analysis user selects the data generation category; the generation element, the associated interaction data and the data generation prompt word are input into the large language model to generate data, and the simulated interaction data is obtained.

[0081] Specifically, if the generation element submitted after the analysis user selects the data generation category is obtained, first, the associated interaction data is queried in the user interaction data according to the query object feature in the generation element, and then the generation element, the associated interaction data and the data generation prompt word are input into the large language model to generate data, and the simulated interaction data is obtained.

[0082] Optionally, the large language model can generate data according to the associated interaction data and the generation object feature in the generation element to obtain simulated interaction data. Optionally, the generation element can include a query object feature and a generation object feature. The query object feature is used to represent the object requirement to be queried, for example, querying users in c1 city who purchased guarantee products in the past 3 to 6 months. The generation object feature is used to represent the generation requirement of the user interaction data to be generated, for example, c2 city is a coastal city, and the user interaction data of simulated users who purchased guarantee products in c2 city in the past 3 to 6 months is generated.

[0083] After obtaining the simulated interaction data, application analysis can be performed based on the simulated interaction data to obtain simulated analysis results. Specifically, after obtaining the simulated interaction data, at least one application analysis question, the simulated interaction data, and an application analysis prompt word can be input into the large language model to perform simulated application analysis and obtain simulated analysis results.

[0084] In addition to first inputting the generation element, the associated interaction data, and the data generation prompt word into the large language model to generate data and obtain simulated interaction data, and then inputting at least one application analysis question, the simulated interaction data, and the application analysis prompt word into the large language model to perform application analysis and obtain simulated analysis results, the generation element, the associated interaction data, at least one application analysis question, and a processing prompt word can also be input into the large language model to generate data and perform application analysis to obtain simulated analysis results. The specific implementation process can refer to the above specific content, which will not be described here. The processing prompt word can be used to instruct the large language model to generate data and perform application analysis.

[0085] In addition, the generation element and the data generation prompt word can also be directly input into the large language model to generate data and obtain simulated interaction data.

[0086] (3) Analysis category In specific implementation, if the analysis manner submitted by the analysis user after selecting the analysis category is obtained, in an optional implementation provided by the embodiment, the analysis manner submitted by the analysis user after selecting the analysis category and an analysis prompt word can be input into the large language model to perform result analysis on the application analysis result, and obtain result analysis data.

[0087] In a specific implementation process, the analysis mode, the application analysis result and the analysis prompt word can be input into the large language model to analyze the application analysis result according to the analysis mode, and obtain result analysis data. In addition, the application analysis result can also be written into the analysis prompt template to obtain the analysis prompt word, and the analysis mode and the analysis prompt word can be input into the large language model to analyze the application analysis result according to the analysis mode, and obtain result analysis data. Alternatively, the application analysis result can not be input, and the analysis prompt word can be used to instruct the large language model to analyze the previously generated application analysis result according to the analysis mode.

[0088] Optionally, the analysis mode includes at least one of the following: descriptive analysis, cross analysis, and factor analysis.

[0089] In addition to analyzing the application analysis result, data analysis can also be performed on the user interaction data or the target interaction data. Specifically, the user interaction data of the application to be analyzed, the analysis mode and the analysis prompt word can be input into the large language model to analyze the user interaction data according to the analysis mode, and obtain data analysis result.

[0090] The above provides a specific process of interaction analysis in the three categories of question and answer category, data generation category and analysis category. The three implementation manners provided above can be selected for implementation according to actual needs in a specific implementation process, or any combination of the implementation manners can also be performed according to actual needs, and in the process of combining the implementation manners provided above, the implementation manners can also be adaptively adjusted according to actual needs. Specifically, the three implementation manners provided above can also be increased, deleted or adjusted in features according to actual needs.

[0091] It should be noted that the interaction analysis performed by selecting the interaction category can be performed based on the application analysis result after obtaining the answer data of each application analysis question as the application analysis result by generating the answer of at least one application analysis question in the application analysis question list based on the target interaction data by the large language model. It can also be performed before obtaining the application analysis result, or it can be performed in parallel with obtaining the application analysis result. This embodiment does not limit the execution order of the interaction analysis and the application analysis.

[0092] To sum up, the application analysis processing method provided in this embodiment, in the process of application analysis processing, obtains the analysis elements submitted by the analysis user for analyzing the application to be analyzed, and then on one hand, inputs the analysis elements and the question generation prompt words into the large language model to generate questions, obtains an application analysis question list for application analysis questions, in order to improve the effectiveness and accuracy of the application analysis questions in the application analysis question list, the application analysis question list can be sent to the analysis user for question confirmation; on the other hand, the target interaction data is determined according to the analysis elements in the user interaction data of the application to be analyzed, so that the analysis data of the user for the application to be analyzed is obtained without interaction with multiple users, the efficiency and convenience of application analysis are improved, and the cost of application analysis is saved; After obtaining the application analysis question list for application analysis questions and the target interaction data for application analysis question answers or application analysis answers, the target interaction data, the application analysis prompt words, the analysis elements and / or at least one application analysis question in the application analysis question list are input into the large language model to generate answers to each application analysis question according to the target interaction data, and the question answers to each application analysis question are obtained as application analysis results, so that the efficiency and convenience of application analysis are improved, and the accuracy and effectiveness of application analysis are ensured; On this basis, at least one interaction category of interaction capability can also be provided, so that the analysis user can perform interaction analysis on the application to be analyzed based on the interaction capability, the flexibility of interaction analysis is improved, and the needs of diversified analysis are met.

[0093] The application analysis processing method provided in this embodiment is further described below with the application of the product analysis scene as an example, as shown in the following Figure 3 The application analysis processing method applied to the product analysis scene includes the following steps.

[0094] Step S302, obtaining the analysis elements submitted by the analysis user for product analysis of the security product.

[0095] Step S304, inputting the analysis elements and the question generation prompt words into the large language model to generate questions, and obtaining an initial question list for product analysis questions.

[0096] Step S306, sending the initial question list to the analysis user.

[0097] In this process, step S306 can be replaced by sending the initial question list to the user terminal of the analysis user.

[0098] Step S308, according to the analysis of the user based on the initial question list after the question editing submitted by the revised confirmation instruction, obtain the revised question list.

[0099] Optionally, the revised question list is obtained after the user analyzes the product analysis question in the initial question list and edits the question.

[0100] Step S310, according to the analysis of the analysis object characteristics in the analysis element, query the target interaction data in the database corresponding to the guarantee product.

[0101] Optionally, the database corresponding to the guarantee product stores the user interaction data of the guarantee product.

[0102] Step S312, input at least one product analysis question in the target interaction data, product analysis prompt word and revised question list into the large language model to generate answers, obtain the question answers of each product analysis question as the product analysis result.

[0103] Step S314, send the product analysis result to the analysis user.

[0104] Step S316, obtain the question submitted by the analysis user based on the product analysis result and select the question and answer category.

[0105] Step S318, input the product analysis result, question and answer generation prompt word into the large language model to generate answers, obtain the answers of the question and send them to the analysis user.

[0106] In addition, steps S316 to S318 can be replaced by obtaining the generation element submitted by the analysis user based on the product analysis result and select the data generation category, inputting the generation element and data generation prompt word into the large language model to generate data according to the generation element, obtaining the simulated interaction data, inputting the simulated interaction data, product analysis prompt word and at least one product analysis question in the revised question list into the large language model to generate answers, and obtaining the question answers of each product analysis question as the simulated analysis result. Among them, inputting the generation element and data generation prompt word into the large language model to generate data and obtain the simulated interaction data can be replaced by querying the associated interaction data in the database of the guarantee product according to the generation element, inputting the generation element, associated interaction data and data generation prompt word into the large language model to generate data according to the generation element and associated interaction data, and obtaining the simulated interaction data; or, it can also be replaced by inputting the generation element, user interaction data in the database and data generation prompt word into the large language model to generate data according to the generation element and user interaction data, and obtaining the simulated interaction data. Further, steps S316 to S318 can also be replaced by obtaining an analysis manner submitted by the analysis user after selecting an analysis category based on the product analysis result, inputting the product analysis result, the analysis manner and the analysis prompt word into the large language model to perform result analysis on the product analysis result, and obtaining result analysis data.

[0107] It should be noted that steps S316 to S318 can also not be performed.

[0108] It should be noted that any one of steps S302 to S318 or a combination of any multiple steps can be combined with any one of steps S202 to S208 or any multiple steps to form a new implementation manner according to the needs of implementation deployment. In addition, any one or more technical features in steps S302 to S318 can also be combined with any one or more technical features provided in steps S202 to S208 to form a new implementation manner according to the actual deployment needs. Alternatively, any one or more technical features in steps S302 to S318 can also be replaced by any one or more technical features provided in steps S202 to S208 to form a new implementation manner according to the actual deployment needs, which will not be described one by one here.

[0109] The application analysis processing device provided in the specification implements, for example: In the above embodiments, an application analysis processing method is provided, and a corresponding application analysis processing device is also provided, which will be described below with reference to the accompanying drawings.

[0110] Reference Figure 4 which shows a schematic diagram of an embodiment of an application analysis processing device provided by the present embodiment.

[0111] Since the device embodiment corresponds to the method embodiment, the description is relatively simple, and the related parts can be referred to the corresponding description of the method embodiment provided above. The device embodiments described below are only illustrative.

[0112] The present embodiment provides an application analysis processing device, which comprises: An analysis element acquisition module 402 is configured to acquire an analysis element for performing application analysis on a to-be-analyzed application; A question generation module 404 is configured to input the analysis element and a question generation prompt word into a large language model to perform question generation, and obtain a list of application analysis questions; A data query module 406 is configured to query target interaction data in user interaction data of the to-be-analyzed application according to the analysis element; The application analysis module 408 is configured to input at least one of the target interaction data, the application analysis prompt word, and the at least one application analysis question in the application analysis question list into a large language model to perform application analysis, and obtain an application analysis result.

[0113] The present specification provides an application analysis processing device, which implements, for example, the following: According to the same technical concept, the present specification one or more embodiments further provide an application analysis processing device for executing the application analysis processing method provided in the present specification, Figure 5 A structural schematic diagram of an application analysis processing device provided in the present specification one or more embodiments.

[0114] The application analysis processing device provided in the present embodiment includes: As Figure 5 As shown, the application analysis processing device can have great differences due to different configurations or performances, and can include one or more processors 501 and memories 502, and the memories 502 can store one or more stored application programs or data. The memory 502 can be temporary storage or persistent storage. The application program stored in the memory 502 can include one or more modules (not shown in the figure), and each module can include a series of computer executable instructions in the application analysis processing device. Further, the processor 501 can be configured to communicate with the memory 502 and execute a series of computer executable instructions in the memory 502 on the application analysis processing device. The application analysis processing device can further include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, one or more keyboards 506, and the like.

[0115] In a specific embodiment, the application analysis processing device includes a memory and one or more programs, wherein one or more programs are stored in the memory, and the one or more programs can include one or more modules, and each module can include a series of computer executable instructions in the application analysis processing device, and the one or more processors configured to execute the one or more programs include computer executable instructions for: Obtaining an analysis element for performing application analysis on the application to be analyzed; Inputting the analysis element and the question generation prompt word into a large language model to generate a question, and obtaining an application analysis question list; According to the analysis element, querying target interaction data in the user interaction data of the application to be analyzed; inputting the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result.

[0116] The computer-readable storage medium provided in the specification implements, for example, the following: According to the application analysis processing method described above, based on the same technical concept, one or more embodiments of the specification further provide a computer-readable storage medium.

[0117] The computer-readable storage medium provided in the embodiment is used to store computer-executable instructions, and the computer-executable instructions, when executed, implement the following processes: obtaining an analysis element for application analysis of the application to be analyzed; inputting the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis question list; querying target interaction data from user interaction data of the application to be analyzed according to the analysis element; inputting the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result.

[0118] It should be noted that the embodiments of the computer-readable storage medium in the specification and the embodiments of the application analysis processing method in the specification are based on the same inventive concept, so the specific implementation of the embodiments can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described herein.

[0119] The computer program product provided in the specification implements, for example, the following: According to the application analysis processing method described above, based on the same technical concept, one or more embodiments of the specification further provide a computer program product.

[0120] A computer program product includes computer programs / instructions, which, when executed by a processor, implement the following steps: obtaining an analysis element for application analysis of the application to be analyzed; inputting the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis question list; querying target interaction data from user interaction data of the application to be analyzed according to the analysis element; inputting the target interaction data, the application analysis prompt word, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result.

[0121] It should be noted that the embodiment of the computer program product in the specification is based on the same inventive concept as the embodiment of the application analysis processing method in the specification, and therefore the specific implementation of the embodiment can be referred to the foregoing implementation of the corresponding method, and the repeated parts will not be described herein.

[0122] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments, such as the device embodiment, the equipment embodiment, and the computer readable storage medium embodiment, which are similar to the method embodiment, and therefore the description is relatively simple. The related content in the device embodiment, the equipment embodiment, and the computer readable storage medium embodiment can be referred to the part of the description of the method embodiment.

[0123] The above describes specific embodiments of the specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims can be performed in a different order than the order in which they are recited and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In some implementations, multiple answer generation and parallel processing are possible or can be advantageous.

[0124] In the 1930s, it was clear to distinguish whether an improvement in a technology was in hardware (e.g., improvement in circuit structure of diodes, transistors, switches, etc.) or in software (e.g., improvement in method flow). However, as technology has evolved, many improvements in method flow today can be considered as direct improvements in hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using hardware entity modules. For example, a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. A digital system is "integrated" on a PLD by the designer programming it, rather than by ordering a chip manufacturer to design and fabricate a custom integrated circuit chip. Moreover, instead of manually fabricating integrated circuit chips, this programming is now mostly implemented using "logic compiler" software, which is similar to software compilers used in program development, and the original code before compilation is written in a specific programming language, which is called a hardware description language (HDL), and there are many such languages, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., and the most commonly used are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should be aware that only a slight logical programming of the method flow in the above-mentioned hardware description languages and programming into an integrated circuit can easily obtain a hardware circuit that implements the logical method flow.

[0125] The controller can be implemented in any suitable way, for example, the controller can take the form of a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to implementing the controller in pure computer readable program code, it is possible to implement the controller in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, etc. to perform the same functions by logically programming the method steps. Such a controller can therefore be considered to be a hardware component, and the means included therein for performing the various functions can also be considered to be structures within the hardware component. Alternatively, or even additionally, the means for performing the various functions can be considered to be both a software module implementing the method and a structure within the hardware component.

[0126] The systems, apparatuses, modules or units illustrated by the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0127] For the sake of description, the above apparatuses are described in functional division and are described respectively. Of course, the functions of the units can be implemented in the same or multiple software and / or hardware when implementing the embodiments of the present specification.

[0128] Those skilled in the art will understand that one or more embodiments of the present specification can be provided as a method, a system or a computer program product. Therefore, one or more embodiments of the present specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present specification can take the form of a computer program product implemented on one or more computer readable storage media (including but not limited to disk storage, CD-ROMs, optical storage devices, etc.) containing computer usable program code.

[0129] The specification is presented with reference to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks.

[0130] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks.

[0131] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus (system) and / or computer program product according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks.

[0132] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0133] The memory can include non-persistent memory, Random Access Memory (RAM), and / or non-volatile memory, such as Read Only Memory (ROM) or flash memory, among others. The memory is an example of computer-readable media.

[0134] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer-readable storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0135] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, such that processes, methods, articles or devices that comprise a list of elements not only include those elements, but also include other elements not expressly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the phrase "comprising at least one" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0136] One or more embodiments of the present specification can be described in the general context of computer-executable instructions being executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types. One or more embodiments of the present specification can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0137] The above only describes the embodiments of the present document and is not intended to limit the present document. For those skilled in the art, the present document can have various changes and variations. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present document shall be included in the scope of claims of the present document.

Claims

1. An application analysis processing method, comprising: obtaining an analysis element for application analysis of an application to be analyzed; inputting the analysis element and a question generation prompt word into a large language model for question generation to obtain a list of application analysis questions; querying target interaction data in user interaction data of the application to be analyzed according to the analysis element; inputting the target interaction data, an application analysis prompt word, and at least one application analysis question in the list of application analysis questions into a large language model for application analysis to obtain an application analysis result.

2. The application analysis processing method of claim 1, wherein the application analysis comprises: reading a statistical analysis question in the at least one application analysis question according to the application analysis prompt word; performing data classification on the target interaction data based on the statistical analysis question to obtain category interaction data under at least one statistical category; generating the application analysis result according to the category interaction data under each statistical category.

3. The application analysis processing method of claim 2, wherein the performing data classification on the target interaction data based on the statistical analysis question to obtain category interaction data under at least one statistical category comprises: reading user evaluation data in the target interaction data according to an evaluation statistical question; performing evaluation classification on the user evaluation data to obtain user evaluation data under each evaluation category.

4. The application analysis processing method of claim 3, wherein the generating the application analysis result according to the category interaction data under each statistical category comprises: calculating an evaluation proportion corresponding to each evaluation category according to a total number of evaluations and a number of user evaluations under each evaluation category; generating an answer to the evaluation statistical question as the application analysis result according to a category identifier and an evaluation proportion corresponding to each evaluation category.

5. The application analysis processing method of claim 1, wherein the application analysis comprises: reading an identification question in the application analysis question according to the application analysis prompt word; reading target dialogue data in the target interaction data based on the identification question; the target dialogue data comprises user dialogue data carrying an identification keyword; performing identification according to the target dialogue data to obtain an identification result as the application analysis result.

6. The application analysis processing method of claim 1, wherein the user interaction data is obtained in the following manner: obtaining user dialogue data related to the application to be analyzed and reading user reference data of a user associated with the user dialogue data; taking the user reference data and the user dialogue data as the user interaction data.

7. The application analysis processing method of claim 1, wherein the querying target interaction data in user interaction data of the application to be analyzed according to the analysis element comprises: reading an analysis object feature in the analysis element; the analysis object feature comprises a time feature and / or a user reference feature; querying target interaction data matching the analysis object feature in the user interaction data.

8. The application analysis processing method of claim 1, after the step of inputting the target interaction data, application analysis prompt words, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result, further comprising: obtaining interaction analysis data submitted by an analysis user after selecting an interaction category; inputting the interaction analysis data and interaction prompt words corresponding to the interaction category into a large language model for response data generation to obtain response data.

9. The application analysis processing method of claim 8, wherein the step of inputting the interaction analysis data and interaction prompt words corresponding to the interaction category into a large language model for response data generation to obtain response data comprises: inputting a question submitted by the analysis user after selecting a question and answer category and answer generation prompt words corresponding to the question and answer category into a large language model for answer generation to obtain an answer to the question.

10. The application analysis processing method of claim 8, wherein the step of inputting the interaction analysis data and interaction prompt words corresponding to the interaction category into a large language model for response data generation to obtain response data comprises: querying associated interaction data in the user interaction data according to a generation element submitted by the analysis user after selecting a data generation category; inputting the generation element, the associated interaction data, and data generation prompt words into a large language model for data generation to obtain simulated interaction data.

11. The application analysis processing method of claim 10, after the step of inputting the generation element, the associated interaction data, and data generation prompt words into a large language model for data generation to obtain simulated interaction data, further comprising: inputting the at least one application analysis question, the simulated interaction data, and the application analysis prompt words into a large language model for application analysis to obtain a simulated analysis result.

12. The application analysis processing method of claim 8, wherein the step of inputting the interaction analysis data and interaction prompt words corresponding to the interaction category into a large language model for response data generation to obtain response data comprises: inputting an analysis method and analysis prompt words submitted by the analysis user after selecting an analysis category into a large language model for result analysis of the application analysis result to obtain result analysis data.

13. The application analysis processing method of claim 1, after the step of inputting the analysis element and question generation prompt words into a large language model for question generation to obtain an application analysis question list, further comprising: sending the application analysis question list to an analysis user; obtaining a revised application analysis question list according to a revision confirmation instruction submitted by the analysis user after editing questions in the application analysis question list; correspondingly, the step of inputting the target interaction data, application analysis prompt words, and at least one application analysis question in the application analysis question list into a large language model for application analysis to obtain an application analysis result comprises: The target interaction data, the application analysis prompt word, and at least one of the modified application analysis problems in the modified application analysis problem list are input into a large language model for application analysis to obtain an application analysis result.

14. The application analysis processing method of claim 1, wherein the question generation comprises: reading a question element in the analysis element according to a question generation prompt word, and generating at least one initial question according to the question element; performing question extension according to the at least one initial question and an analysis category in the analysis element to obtain at least one associated question; generating the application analysis problem list according to the at least one initial question and the at least one associated question.

15. An application analysis processing apparatus, comprising: an analysis element acquisition module configured to acquire an analysis element for performing application analysis on a to-be-analyzed application; a question generation module configured to input the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis problem list; a data query module configured to query target interaction data from user interaction data of the to-be-analyzed application according to the analysis element; an application analysis module configured to input the target interaction data, an application analysis prompt word, and at least one of the application analysis problems in the application analysis problem list into a large language model for application analysis to obtain an application analysis result.

16. An application analysis processing device, comprising: a processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to: acquire an analysis element for performing application analysis on a to-be-analyzed application; input the analysis element and a question generation prompt word into a large language model for question generation to obtain an application analysis problem list; query target interaction data from user interaction data of the to-be-analyzed application according to the analysis element; input the target interaction data, an application analysis prompt word, and at least one of the application analysis problems in the application analysis problem list into a large language model for application analysis to obtain an application analysis result.

17. A computer-readable storage medium for storing computer-executable instructions that, when executed, implement the steps of the method of claim 1.

Images