User feedback information analysis method, device, equipment, medium and program product

Abstract

This application provides a user feedback information analysis method, apparatus, device, medium, and program product, which can be applied to the field of artificial intelligence technology. The method includes: determining the correspondence between user feedback information and user operation logs; ensuring that any user feedback information and its corresponding user operation log are determined for the same user; the user operation log is used to represent a sequence of operation information; for any user feedback information, determining operation information associated with the user feedback information from the sequence of operation information represented by the corresponding user operation log, and determining user improvement needs based on the user feedback information and the determined operation information; and determining overall improvement needs based on the user improvement needs determined for different user feedback information.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to a method, apparatus, device, medium, and program product for analyzing user feedback information. Background Technology

[0002] With the widespread adoption of digitalization, users can use a variety of services to meet their needs. For example, they can use video application services to watch videos or use financial service applications to conduct financial transactions.

[0003] To better serve users, it is often necessary to collect and analyze user feedback. However, currently, this analysis is typically done manually, which is inefficient. Summary of the Invention

[0004] In view of the above problems, this application provides a user feedback information analysis method, apparatus, equipment, medium and program product to improve the efficiency of user feedback information analysis.

[0005] According to the first aspect of this application, a method for analyzing user feedback information is provided, comprising:

[0006] Determine the correspondence between user feedback information and user operation logs; each user feedback information and its corresponding user operation log are determined for the same user; the user operation log is used to represent the sequence of operation information.

[0007] For any user feedback information, determine the operation information associated with the user feedback information from the operation information sequence represented by the corresponding user operation log, and determine the user's improvement needs based on the user feedback information and the determined operation information.

[0008] Based on the user improvement needs identified from different user feedback, the overall improvement needs are determined.

[0009] Optionally, determining the correspondence between user feedback information and user operation logs includes: obtaining user authorization to use user operation logs for analyzing user feedback information; and, if user authorization is obtained, determining the correspondence between user feedback information and user operation logs.

[0010] Optionally, determining the operation information associated with any user feedback information from the operation information sequence represented by the corresponding user operation log includes: for any user feedback information, inputting the user feedback information and the corresponding user operation log into a pre-trained analysis model, and determining the operation information associated with the user feedback information predicted by the analysis model from the operation information sequence represented by the corresponding user operation log; the analysis model is used to extract features from the input user feedback information and user operation log through different feature extraction modules.

[0011] Optionally, determining user improvement needs based on the user feedback information and the determined operation information includes: determining user improvement needs based on the user feedback information and the determined operation information, and the degree of impact of the determined user improvement needs on user satisfaction; determining overall improvement needs based on the user improvement needs determined for different user feedback information includes: determining overall improvement needs based on the user improvement needs determined for different user feedback information, and the degree of impact of the user improvement needs on user satisfaction.

[0012] Optionally, the method further includes: determining user emotion information for any user feedback information; determining user improvement needs based on the targeted user feedback information and the determined operation information includes: determining user improvement needs based on the targeted user feedback information, the determined operation information, and the determined user emotion information; determining overall improvement needs based on the user improvement needs determined for different user feedback information includes: determining overall improvement needs based on the user improvement needs determined for different user feedback information and the user emotion information determined for different user feedback information.

[0013] Optionally, the method further includes: determining user emotional information for any user feedback information; the step of determining user improvement needs and the degree of influence of the determined user improvement needs on user satisfaction based on the user feedback information and the determined operation information includes: determining user improvement needs and the degree of influence of the determined user improvement needs on user satisfaction based on the user feedback information, the determined operation information, and the determined user emotional information.

[0014] Optionally, determining the overall improvement requirements based on user improvement requirements identified from different user feedback information includes: determining the priority of user improvement requirements based on user improvement requirements identified from different user feedback information; and identifying user improvement requirements with a priority greater than a preset priority threshold as overall improvement requirements.

[0015] Optionally, the method for determining the user feedback information includes: for any business process, when the business process has ended, pushing a preset set of questions to the user; determining the user feedback information based on the user's response to the preset set of questions; the preset set of questions is determined based on the business process in question.

[0016] A second aspect of this application provides a user feedback information analysis device, comprising:

[0017] The correspondence unit is used to determine the correspondence between user feedback information and user operation logs; any user feedback information and its corresponding user operation log are determined for the same user; the user operation log is used to represent the sequence of operation information.

[0018] The association unit is used to determine the operation information associated with the user feedback information from the operation information sequence represented by the corresponding user operation log, and to determine the user's improvement needs based on the user feedback information and the determined operation information.

[0019] The requirement unit is used to determine the overall improvement requirements based on user improvement requirements identified from different user feedback information.

[0020] Optionally, the correspondence unit is used to: obtain the user's authorization to use the user operation log to analyze user feedback information; and, if the user's authorization to use the user operation log to analyze user feedback information is obtained, determine the correspondence between the user feedback information and the user operation log.

[0021] Optionally, the association unit is used to: input the target user feedback information and the corresponding user operation log into a pre-trained analysis model for any user feedback information, and determine the operation information associated with the target user feedback information predicted by the analysis model from the operation information sequence represented by the corresponding user operation log; the analysis model is used to extract features from the input user feedback information and user operation log through different feature extraction modules respectively.

[0022] Optionally, the association unit is used to: determine user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction based on the user feedback information and the determined operation information; the requirement unit is used to: determine overall improvement needs based on the user improvement needs determined for different user feedback information and the degree of impact of the user improvement needs on user satisfaction.

[0023] Optionally, the device further includes: an emotion unit, configured to: determine user emotion information in response to any user feedback information; an association unit, configured to: determine user improvement needs based on the user feedback information, the determined operation information, and the determined user emotion information; and a requirement unit, configured to: determine overall improvement requirements based on the user improvement requirements determined for different user feedback information and the user emotion information determined for different user feedback information.

[0024] Optionally, the device further includes: an emotion unit, configured to: determine user emotion information in response to any user feedback information; and an association unit, configured to: determine user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction based on the user feedback information, the determined operation information, and the determined user emotion information.

[0025] Optionally, the requirement unit is used to: determine the requirement priority of user improvement requirements based on user improvement requirements identified in response to different user feedback information; and determine user improvement requirements with a requirement priority greater than a preset priority threshold as overall improvement requirements.

[0026] Optionally, the method for determining the user feedback information includes: for any business process, when the business process has ended, pushing a preset set of questions to the user; determining the user feedback information based on the user's response to the preset set of questions; the preset set of questions is determined based on the business process in question.

[0027] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0028] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0029] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0030] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0031] Figure 1 This illustration schematically depicts an application scenario of a user feedback information analysis method according to an embodiment of this application.

[0032] Figure 2 A flowchart illustrating a user feedback information analysis method according to an embodiment of this application is shown schematically.

[0033] Figure 3 This schematically illustrates a structural block diagram of a user feedback information analysis device according to an embodiment of this application; and

[0034] Figure 4 A block diagram schematically illustrates an electronic device suitable for implementing a user feedback information analysis method according to an embodiment of this application. Detailed Implementation

[0035] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0036] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0037] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0038] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0039] With the widespread adoption of digitalization, users can use a variety of services to meet their needs. For example, they can use video application services to watch videos or use financial service applications to conduct financial transactions.

[0040] To better serve users, it is often necessary to collect and analyze user feedback. However, currently, this analysis is typically done manually, which is inefficient.

[0041] This application provides a method for analyzing user feedback information. In this method, user feedback information can be automatically analyzed. Specifically, it can combine user feedback information with corresponding user operation logs to automatically analyze and determine relevant user improvement needs. Based on the user improvement needs determined from different user feedback information, overall improvement needs can be determined.

[0042] In a specific example, user feedback could be in text form, such as "the operation is cumbersome." This feedback, combined with the user's operation logs, allows analysis of the user's time-consuming operations. For example, if the authentication operation takes longer than average, the corresponding user improvement requirement can be determined as "reducing the operational burden for authentication." Different user feedback from different users can be analyzed in conjunction with their operation logs to determine corresponding user improvement requirements. This allows for the synthesis of these diverse user needs to determine the overall improvement requirement. For instance, if multiple users have the user improvement requirement of "reducing the burden for operation A," while only one user has the requirement of "reducing the burden for operation B," then "reducing the burden for operation A" can be identified as the overall improvement requirement based on the overall user needs.

[0043] This method can improve the efficiency of user feedback analysis by automatically analyzing user feedback information, and can also improve the accuracy of user feedback analysis by combining it with user operation logs, thereby improving the accuracy and comprehensiveness of user improvement needs and overall improvement needs.

[0044] Understandably, analyzing users' overall improvement needs makes it easier to develop based on these needs, distinguish the development priorities of different needs, thereby improving development efficiency and the effectiveness of business improvements, better meeting users' actual needs, and enhancing user experience and satisfaction.

[0045] It should be noted that the methods and apparatus disclosed in the embodiments of this application can be used in the field of artificial intelligence technology, and also in the field of fintech. For example, user feedback information in the financial field can be analyzed using the above methods; it can also be applied to any field other than fintech, for example, analyzing user feedback information in other fields using the above methods. The application fields of the methods and apparatus disclosed in the embodiments of this application are not limited.

[0046] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user feedback information, user operation logs, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0047] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this application all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0048] Figure 1 The illustration depicts an application scenario of a user feedback information analysis method according to an embodiment of this application. For example... Figure 1As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc. Users can use the first terminal device 101, the second terminal device 102, or the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, or the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (only examples). The first terminal device 101, the second terminal device 102, or the third terminal device 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers. The server 105 can be a server providing various services, such as a backend management server (for example only) that supports websites browsed by users using the first terminal device 101, the second terminal device 102, or the third terminal device 103. The backend management server can analyze and process received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0049] It is understood that users can perform automatic analysis of user feedback information locally on their terminal devices, thereby executing the user feedback information analysis method provided in this application embodiment through the terminal devices. Users can also perform automatic analysis of user feedback information on the server 105, and execute the user feedback information analysis method provided in this application embodiment through the server 105.

[0050] It should be noted that the user feedback information analysis method provided in this application embodiment can generally be executed by server 105, first terminal device 101, second terminal device 102, or third terminal device 103. Correspondingly, the user feedback information analysis device provided in this application embodiment can generally be located in server 105, first terminal device 101, second terminal device 102, or third terminal device 103. The user feedback information analysis method provided in this application embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with first terminal device 101, second terminal device 102, third terminal device 103, and / or server 105. Correspondingly, the user feedback information analysis device provided in this application embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with first terminal device 101, second terminal device 102, third terminal device 103, and / or server 105. It should be understood that... Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0051] Figure 2 A flowchart illustrating a user feedback information analysis method according to an embodiment of this application is shown schematically.

[0052] like Figure 2 As shown, the method flow of this embodiment may include operations S210 to S230. This application embodiment does not limit the specific executing entity; it can be any electronic device or any software application, such as a user terminal, server, or so on.

[0053] In operation S210, the correspondence between user feedback information and user operation log is determined; any user feedback information and the corresponding user operation log are determined for the same user; the user operation log is used to represent the sequence of operation information.

[0054] In operation S220, for any user feedback information, the operation information associated with the user feedback information is determined from the operation information sequence represented by the corresponding user operation log, and the user's improvement requirements are determined based on the user feedback information and the determined operation information.

[0055] In operation S230, the overall improvement requirements are determined based on the user improvement requirements identified in response to different user feedback information.

[0056] This method can improve the efficiency of user feedback analysis by automatically analyzing user feedback information.

[0057] This methodology can also improve the accuracy of user feedback analysis by combining user operation logs with user feedback information, thereby enhancing the accuracy and comprehensiveness of user improvement requests and overall improvement requirements.

[0058] The embodiments of this application are not limited to user feedback information. Optionally, user feedback information can be used to characterize the feedback provided by the user, specifically feedback on business processes or product usage. For example, user feedback information can be in text form such as "The xx service is rather cumbersome," or in voice form obtained through a telephone interview such as "The buttons for the xx service are not sensitive," etc.

[0059] The embodiments of this application do not limit the method for determining user feedback information. Optionally, user feedback information sent by the user can be obtained directly, for example, the user sends user feedback information through a client or terminal. Optionally, user feedback information can also be extracted based on information provided by the user. For example, the user can provide feedback by filling out a questionnaire, and user feedback information can be extracted based on the user's questionnaire results.

[0060] In a specific example, user questionnaire results can be obtained, which may include satisfaction with feature A, suggestions for improvement of feature B, etc., thereby extracting one or more user feedback information.

[0061] The embodiments of this application do not limit the specific form of user feedback information. Optionally, user feedback information may be in the form of text, images, voice, video, etc.

[0062] In one optional embodiment, user feedback information may be unclear. Therefore, user operation logs can be used to analyze the user's actual operation, thereby determining the user's actual feedback and further identifying corresponding improvement needs. In a specific example, user feedback may simply be the text "The process is cumbersome," making it difficult to pinpoint the corresponding process. Therefore, by analyzing user operation logs, the user's actual operation can be identified as the "transfer process," thus more accurately determining the user's actual feedback as "The transfer process is cumbersome," and further identifying the corresponding improvement need as "Simplifying the transfer process."

[0063] The embodiments of this application do not limit user operation logs. Optionally, a user operation log can be a log recorded for user operations, or an operation log extracted from user logs. Optionally, a user operation log can be used to represent a sequence of user operation information, which may include one or more operation information items ordered by time. The embodiments of this application do not limit the specific content and form of the user operation log. Optionally, the user operation log may include operation information and corresponding time information, whereby the corresponding time information can be used to represent the time when the operation represented by the operation information occurred, thereby allowing the operation information to be ordered according to the time information to obtain an operation information sequence.

[0064] The embodiments of this application do not limit the operation information. Optionally, operation information can be used to characterize user operations. For example, when a user clicks a button, a corresponding button click log can be generated in the user operation log to characterize the user's button click operation. The embodiments of this application do not limit the specific content and form of the operation information. Optionally, the operation information may include information such as the operation's attributes, the business to which the operation belongs, the type of the operation, the time the operation was generated, the duration of the operation, the waiting time of the operation, the response speed of the operation, and the number of operation steps. It is understood that the operation information can be used to subsequently analyze user feedback information and determine user improvement needs. For example, if the user feedback information is "The process of business A is too long," then the operation information related to business A associated with this user feedback information can be determined, thereby analyzing the time-consuming operations among the various operation information to further determine the user's improvement needs.

[0065] The embodiments of this application do not limit the method of obtaining user feedback information and user operation logs, nor do they limit the method of determining them. Optionally, user feedback information can be collected from users, and with the user's authorization, the corresponding user's user operation logs can be obtained for analysis of the user feedback information.

[0066] The embodiments of this application do not limit the method for determining the correspondence between user feedback information and user operation logs. Optionally, any user feedback information and its corresponding user operation log can be determined for the same user. It is understood that for any user feedback information, the corresponding user operation log can be obtained for the user who provided the feedback information. It is understood that a single user can provide multiple user feedback information, and thus the user operation log for each user feedback information can be obtained separately as the corresponding user operation log. Optionally, a single user feedback information can correspond to one or more user operation logs; a single user operation log can correspond to one or more user feedback information.

[0067] The embodiments of this application do not limit the specific scope of obtaining user operation logs. Optionally, user operation logs of the same user prior to the time when user feedback information was provided can be obtained, specifically, user operation logs of the same user within a preset time period prior to the time when the feedback information was provided. In the embodiments of this application, user information such as user operation logs is obtained with the user's authorization.

[0068] Optionally, determining the correspondence between user feedback information and user operation logs can specifically involve: obtaining user authorization to use user operation logs for analyzing user feedback information; and, if user authorization is obtained, determining the correspondence between user feedback information and user operation logs. This embodiment can improve the security of user information by determining the correspondence between user feedback information and user operation logs after obtaining user authorization, which is then used for subsequent analysis of user feedback information.

[0069] In embodiments of this application, user consent or authorization may be obtained before acquiring user information (e.g., user operation logs and / or user feedback information). For example, before operation S210, a request may be sent to the user to acquire user information (e.g., user operation logs and / or user feedback information). If the user consents or authorizes the acquisition of user information (e.g., user operation logs and / or user feedback information), operation S210 is performed.

[0070] The embodiments of this application do not limit the number of correspondences determined in S210. Optionally, it may be determining the correspondence between M groups of user feedback information and user operation logs. M is a positive integer. Alternatively, it may be determining the correspondence between one or more groups of user feedback information and user operation logs. By determining multiple sets of correspondences, it is convenient to comprehensively determine the overall improvement requirements based on the user improvement needs identified for different user feedback information.

[0071] The embodiments of this application do not limit the method for determining the association between user feedback information and the operation information represented by the corresponding user operation log. In an optional embodiment, determining the operation information associated with the user feedback information may specifically involve determining the operation information corresponding to the business operation involved in the user feedback information. For example, if the user feedback information is "Business A is quite complicated", then the operation information related to business A in the user operation log can be determined as the operation information associated with the user feedback information.

[0072] Optionally, based on a large language model, for user feedback information, the associated operation information can be determined from the operation information sequence represented by the corresponding user operation log. This association can be used to represent the relationship between the business operations involved in the user feedback information and the business operations represented by the operation information. Alternatively, the associated operation information can also be determined by determining the feature similarity between the user feedback information and the operation information.

[0073] Optionally, for any user feedback information, the operation information associated with the user feedback information is determined from the operation information sequence represented by the corresponding user operation log. Specifically, for any user feedback information, the user feedback information and the corresponding user operation log are input into a pre-trained analysis model to determine the operation information associated with the user feedback information predicted by the analysis model from the operation information sequence represented by the corresponding user operation log. The analysis model can be used to extract features from the input user feedback information and user operation log separately through different feature extraction modules. This embodiment can improve the feature extraction effect of user feedback information and user operation log by extracting features from user feedback information and user operation log separately through the analysis model, thereby improving the accuracy of user feedback information analysis and the accuracy of the association between user feedback information and operation information.

[0074] The embodiments of this application do not limit the specific structure and operation process of the analysis model. Optionally, the analysis model may include two different feature extraction modules, which can extract features from the input user feedback information and user operation logs respectively, and further determine the correlation between the user feedback information and the operation information in the user operation logs based on the extracted features. The embodiments of this application do not limit the input format of the analysis model. It is understood that multiple user feedback information and multiple user operation logs can be input into the analysis model to determine the above-mentioned correlation, or a single user feedback information and multiple user operation logs can be input into the analysis model to determine the above-mentioned correlation, or multiple user feedback information and a single user operation log can be input into the analysis model to determine the above-mentioned correlation. Optionally, the analysis model can further determine the above-mentioned correlation based on the features extracted from the input user feedback information and the features extracted from the input user operation logs, according to an attention mechanism. Specifically, the user feedback information can be treated as a sequence, and the user operation logs or the sequence of operation information represented by the user operation logs can also be treated as a sequence, and input into the attention mechanism to determine the above-mentioned correlation.

[0075] The embodiments of this application are not limited to a specific method for determining user improvement needs based on user feedback information and associated operation information. Optionally, user problems in actual operation can be analyzed based on user feedback information and the operation information associated with the user feedback information. Specifically, this may include business positioning information where the user has feedback, such as the part of the "business process" or "operation" where the user has feedback, and may also include the type of user feedback, such as cumbersome operation, insensitive, inconvenient, or having too many steps.

[0076] The embodiments of this application do not limit the specific form and content of user improvement requests. Optionally, user improvement requests may include the business positioning information to be improved, as well as the type of improvement information. For example, a user improvement request may represent "improving the simplicity of business A", or "improving the response speed of authentication operations in business B", and so on.

[0077] Accordingly, when determining user improvement needs based on user feedback and associated operational information, the specific business positioning information and type of improvement required can be determined from the user feedback and associated operational information, thus identifying the user's improvement needs. Optionally, user feedback and associated operational information can be analyzed comprehensively. For example, if the user feedback explicitly mentions a certain business, the business positioning information can be directly determined based on the user feedback. If the user feedback does not mention a business but only mentions "cumbersome operation," the business positioning information can be determined by combining the business operations with many steps in the associated operational information. For example, if the user feedback mentions that the type of improvement needed is "improving operational sensitivity," the type of improvement needed can be directly determined based on the user feedback; if the user feedback does not mention the type of improvement but only mentions "Business B," the characteristics of Business B operations can be analyzed from the associated operational information. If the characteristic of Business B operations is a large number of steps, the type of improvement needed is to reduce the number of steps; if the characteristic of Business B operations is long processing time, the type of improvement needed is to reduce operation time. The business positioning information is business B, and improvements need to be made for business B.

[0078] It is understood that user improvement requests can also take other forms or have other content, and the corresponding determination methods are not limited. The above embodiments are merely illustrative examples.

[0079] Optionally, user improvement needs can be determined based on a large language model, using user feedback information and associated operation information. Alternatively, a first candidate improvement need can be determined based on user feedback information, and a second candidate improvement need can be determined based on associated operation information. The first and second candidate improvement needs can then be combined to determine the final user improvement need. Specifically, both the first and second candidate improvement needs can be identified as user improvement needs, or overlapping needs between the first and second candidate improvement needs can be identified as user improvement needs. Specifically, determining the first candidate improvement need based on user feedback information can be done through methods such as semantic understanding. Specifically, determining the second candidate improvement need based on associated operation information can be done by analyzing the characteristics of the operation information to further identify the corresponding second candidate improvement need.

[0080] In a specific example, user improvement requests can have fixed types, thus facilitating the standardization of their form and content. For instance, the business positioning information in a user improvement request could have 10 selectable scenarios, from business 1 to business 10, and the type of improvement required could have fixed options such as "simplified operation," "improved sensitivity," and "improved response speed."

[0081] Therefore, alternatively, user improvement needs can be set as sample labels through model training, and the corresponding user feedback information and associated operation information can be used to determine sample features to form training samples for model training. The trained model can be used to predict and determine user improvement needs based on user feedback information and associated operation information.

[0082] The embodiments of this application are not limited to a specific method for determining the overall improvement requirements based on user improvement needs. Optionally, the overall improvement requirements can be determined by filtering the multiple determined user improvement requirements; the overall improvement requirements can also be determined by merging the multiple determined user improvement requirements; or the overall improvement requirements can be determined by statistically analyzing the multiple determined user improvement requirements and based on the statistical values.

[0083] For example, among the identified multiple user improvement requests, the more important ones can be further filtered out and identified as overall improvement requests. Alternatively, among the identified multiple user improvement requests, the number of identical user improvement requests can be counted, and user improvement requests with a number greater than a preset threshold can be identified as overall improvement requests. Or, user improvement requests with a percentage greater than a preset percentage threshold can be identified as overall improvement requests.

[0084] In an alternative embodiment, the accuracy of overall improvement requirements can be improved by incorporating the degree of impact of user improvement requests on user satisfaction.

[0085] Optionally, user improvement needs are determined based on the targeted user feedback information and the determined operational information. Specifically, this can involve determining user improvement needs and their impact on user satisfaction based on the targeted user feedback information and the determined operational information. Correspondingly, overall improvement needs are determined based on the user improvement needs determined for different user feedback information. Specifically, this can involve determining overall improvement needs based on the user improvement needs determined for different user feedback information and their impact on user satisfaction. This embodiment can improve the accuracy of overall improvement needs and facilitate improved user satisfaction by determining overall improvement needs based on their impact on user satisfaction.

[0086] The embodiments of this application do not limit the specific form of the impact of user improvement requests on user satisfaction. Optionally, the impact of user improvement requests on user satisfaction may specifically include: the degree of decrease in user satisfaction when user improvement requests are not implemented, and / or the degree of increase in user satisfaction when user improvement requests are implemented.

[0087] The embodiments of this application do not limit the specific method for determining the overall improvement requirements based on the degree of impact of user improvement requirements on user satisfaction. Optionally, user improvement requirements that result in a decrease in user satisfaction greater than a preset decrease when user improvement requirements are not implemented, and / or user improvement requirements that result in an increase in user satisfaction greater than a preset increase when user improvement requirements are implemented, can be determined as overall improvement requirements. Optionally, user improvement requirements whose impact on user satisfaction is greater than a preset impact can also be determined as overall improvement requirements.

[0088] The embodiments of this application do not limit the specific method for determining the impact of user improvement requests on user satisfaction. Optionally, the impact of user improvement requests on user satisfaction can be analyzed based on user feedback information and associated operational information. Specifically, a large language model or other pre-trained models can be used to predict and determine the impact of user improvement requests on user satisfaction.

[0089] In a specific example, if user feedback repeatedly mentions improvements to a particular service or contains strong emotional responses, it can be determined that the corresponding user improvement request has a significant impact on user satisfaction. Similarly, if the associated operation information includes multiple failed attempts related to a particular service, it can be determined that the corresponding user improvement request has a significant impact on user satisfaction.

[0090] In an optional embodiment, user emotion information can also be introduced to determine user improvement needs or overall improvement needs, which can improve the accuracy of improvement needs. The embodiments of this application do not limit the method of determining user emotions; specifically, it can be determined based on user feedback information. A large language model can be used to determine the emotion information in user feedback information, or other pre-trained models can be used to predict the emotion information in user feedback information.

[0091] Optionally, user emotion information can be determined for any user feedback information; based on the targeted user feedback information and the determined operation information, user improvement needs can be determined. Specifically, this can be done by determining user improvement needs based on the targeted user feedback information, the determined operation information, and the determined user emotion information; based on the user improvement needs determined for different user feedback information, overall improvement needs can be determined. Specifically, this can be done by determining overall improvement needs based on the user improvement needs determined for different user feedback information and the user emotion information determined for different user feedback information. This embodiment can improve the accuracy of improvement needs by determining improvement needs based on user emotion information.

[0092] The embodiments of this application do not limit the specific method of determining improvement needs (user improvement needs or overall improvement needs) based on user emotional information. Optionally, improvement needs involved in user feedback information can be filtered based on user emotional information, or the degree of impact of different user improvement needs on user satisfaction or the priority of user improvement needs can be determined based on user emotional information, thereby determining the overall improvement needs.

[0093] In one optional embodiment, the impact of user improvement requests on user satisfaction can be determined by incorporating user emotional information. It is understood that, in a specific example, if user feedback indicates that the user's emotional information represents anger, it can be determined that the impact of user improvement requests on user satisfaction is relatively large. Determining the impact of user improvement requests on user satisfaction based on user emotional information can improve the accuracy of this assessment.

[0094] Therefore, optionally, user emotional information can be determined for any user feedback information; based on the targeted user feedback information and the determined operational information, user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction can be determined. Specifically, this can be done by determining user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction based on the targeted user feedback information, the determined operational information, and the determined user emotional information. This embodiment can improve the accuracy of determining the degree of impact of user improvement needs on user satisfaction by using user emotional information.

[0095] The embodiments of this application do not limit the specific method of determining overall improvement requirements. Optionally, determining overall improvement requirements based on user improvement requirements identified from different user feedback information can specifically involve: determining the priority of user improvement requirements based on the user improvement requirements identified from different user feedback information; and identifying user improvement requirements with a priority greater than a preset priority threshold as overall improvement requirements. This embodiment can improve the accuracy of overall improvement requirements by filtering user improvement requirements based on their priority.

[0096] The embodiments of this application do not limit the specific method for determining the priority of needs. Optionally, the priority of needs can be used to characterize the degree of impact of user improvement needs on user satisfaction, or it can be used to characterize the urgency of user improvement needs, etc.

[0097] In a specific example, requirement priority can be positively correlated with the degree of decrease in user satisfaction when user improvement requirements are not met; the greater the decrease in user satisfaction when user improvement requirements are not met, the higher the requirement priority. Requirement priority can also be positively correlated with the degree of increase in user satisfaction when user improvement requirements are met; the greater the increase in user satisfaction when user improvement requirements are met, the higher the requirement priority.

[0098] The embodiments of this application do not limit the content and form of specific overall improvement requirements. Optionally, overall improvement requirements can be overall improvement requirements summarized from different user feedback information from multiple users. For example, by collecting a large amount of user feedback information, the overall improvement requirements of a user group can be summarized and determined. Overall improvement requirements can be used to characterize the improvement needs of a user group. Optionally, overall improvement requirements can be determined by filtering from the determined user improvement requirements, specifically including one or more selected user improvement requirements. Overall improvement requirements can also be improvement requirements determined by comprehensively considering the determined user improvement requirements. For example, by summarizing the determined user improvement requirements using a large language model, the overall improvement requirements can include the summary content of the determined user improvement requirements. For example, 80% of users proposed user improvement requirement xx, and 30% of users proposed user improvement requirement yy.

[0099] The embodiments of this application do not limit the subsequent processing method of the overall improvement requirements. Optionally, a subsequent development plan or development priority can be determined based on the overall improvement requirements, thereby facilitating the allocation of development resources and implementing higher-priority improvement requirements to improve user satisfaction and user experience. For example, the overall improvement requirements may include: 80% of users submitted user xx improvement requirements, and 30% of users submitted user yy improvement requirements, among which user xx improvement requirements have a greater impact on user satisfaction. Therefore, when development resources are limited, user xx improvement requirements can be implemented first.

[0100] The embodiments of this application do not limit the specific form and content of user feedback information, nor do they limit the method of determining user feedback information. Optionally, user feedback information can be in the form of text, voice, video, image, etc. For example, user feedback information can be a text entered by the user such as "The operation process is too cumbersome," or it can be voice or video recorded by the user, etc.

[0101] Optionally, user feedback information can be determined based on the answers to user surveys, or it can be determined directly from the information entered by the user.

[0102] Optionally, the method for determining user feedback information includes: for any business process, upon completion of the business process, pushing a preset set of questions to the user; determining user feedback information based on the user's responses to the preset set of questions; the preset set of questions is determined based on the business process in question. This embodiment can determine the set of questions based on the completed business process, thereby determining user feedback information based on the user's responses, which can improve the real-time performance and accuracy of user feedback information.

[0103] The embodiments of this application do not limit the specific method for determining the preset question set. Optionally, it can be based on a preset question template, combined with information from the targeted business process, to generate corresponding questions and add them to the preset question set. For example, if the targeted business process is abnormally terminated, the question "Abnormal termination of the xx business process was detected, what is the reason?" can be generated.

[0104] Optionally, the pre-set question set can be in the form of a questionnaire, which makes it convenient to collect user feedback information through the questionnaire.

[0105] Optionally, the termination of the business process can be either a failure or a success.

[0106] Optionally, based on the target business process, the actual operation status or characteristics of the user can be analyzed through relevant user operation logs, thereby identifying corresponding issues to add to a preset issue set. For example, if the user operation logs indicate a user operation failure, the cause of the failure or the specific business context of the failure can be determined, thus identifying the corresponding issue and facilitating more accurate acquisition of real-time user feedback information.

[0107] Optionally, relevant questions can be generated and added to a preset question set by combining users' historical feedback information or their historical questionnaire results. For example, based on users' historical feedback information, a corresponding question could be generated: "The software has been updated in response to the previous feedback. How was your experience?"

[0108] For ease of understanding, this application also provides an application embodiment.

[0109] This embodiment relates to information system management and artificial intelligence technology in the field of financial technology. It proposes an intelligent method for predicting the return on investment based on user behavior data and real-time journey analysis, which is particularly suitable for demand verification and optimization throughout the entire life cycle of financial products.

[0110] The purpose of this embodiment is to propose a methodology for predicting product launch effectiveness based on user survey questionnaires, aiming to solve problems such as data lag, strong subjectivity, and low prediction accuracy. This embodiment enables a more scientific, objective, and accurate prediction of product launch effectiveness, providing strong support for product optimization and launch decisions. Specifically, this embodiment deeply analyzes user survey questionnaire data, extracts user feedback on the product, and combines market trends, competitive landscape, and other factors to construct a complete product launch effectiveness prediction model. This model comprehensively considers multiple dimensions such as user needs, market changes, and enterprise resources, thereby achieving a comprehensive prediction of product launch effectiveness. Furthermore, this embodiment emphasizes improving prediction accuracy and practicality. By continuously optimizing the prediction algorithm and model parameters, this embodiment can achieve accurate prediction of product launch effectiveness, providing a more reliable basis for decision-making. Simultaneously, this embodiment also provides easy-to-use data processing and analysis tools for convenient and rapid implementation and application.

[0111] To more scientifically, objectively, and accurately predict product launch effects and provide strong support for product optimization and launch decisions, this embodiment proposes an intelligent launch effect generation method and system based on user surveys. The system consists of four core subsystems, which are tightly coupled through data flow to form a closed-loop optimization process: 1. User behavior analysis subsystem; 2. Dynamic questionnaire generation subsystem; 3. Launch effect prediction subsystem; 4. Attribution analysis and suggestion generation subsystem.

[0112] The detailed description of this embodiment is as follows:

[0113] Process 1: Customer segmentation based on financial scenario behavior paths (user behavior analysis subsystem).

[0114] Objective: To address the issues of coarse granularity and lack of dynamic behavioral analysis in traditional customer segmentation within financial scenarios.

[0115] Step 1.1: Data input and preprocessing.

[0116] Inputs: Real-time user behavior data: function usage time (e.g., payment time, page dwell time), operation path (e.g., "Homepage → Transfer → Verification → Complete"), error logs (e.g., number of identity verification failures); Financial attribute data: product type held (credit card, wealth management, loan), risk preference tags. In this embodiment, all user data is obtained and used only after user authorization.

[0117] Operations: Data cleaning, removing invalid operations (such as accidental page redirects), and filling in missing values ​​(such as inferring based on held products when users have not filled in their risk preferences); Feature engineering, defining key indicators for financial scenarios: Operational efficiency indicators: average number of steps to complete a transfer, browsing depth of wealth management products (page levels); Risk sensitivity indicators: closing speed of risk warning pop-ups, click frequency of high-risk products; Error tolerance: number of retries for errors in the same process.

[0118] Output: Structured behavioral feature matrix (user ID × feature dimension).

[0119] Step 1.2: Customer segmentation algorithm.

[0120] Input: Behavioral feature matrix.

[0121] Operation: Identify core user groups through cluster analysis or classification algorithms: users who frequently complete transfers and quickly browse financial products are marked as "experienced users"; identify peripheral user groups: users whose operation paths are scattered and whose processes are frequently interrupted are marked as "novice users"; Dynamic update: update cluster results every hour to adapt to real-time behavioral changes.

[0122] Output: Customer type label (beginner / expert), coordinates of behavior cluster centers.

[0123] Example: User A's behavioral characteristics: [Number of transfer steps = 3, depth of financial browsing = 5, number of error retries = 0] → identified as "experienced user"; User B's behavioral characteristics: [Number of transfer steps = 6, depth of financial browsing = 2, number of error retries = 3] → identified as "novice user".

[0124] Process 2: Dynamic questionnaire generation triggered by user journey (dynamic questionnaire generation subsystem).

[0125] Objective: To address the issues of traditional questionnaires being static and disconnected from real-time user interaction scenarios, to achieve high-precision dynamic questionnaire generation based on user behavior context, and to improve data collection efficiency and quality through algorithm optimization.

[0126] Step 2.1: Multimodal event perception and dynamic rule engine.

[0127] Inputs: Real-time user operation logs: operation path (e.g., "staying on the financial management page → exiting the risk assessment"), function usage duration, error type (e.g., identity verification failure); user classification tags: novice user / experienced user; historical interaction data: user's past questionnaire response records, question jump path.

[0128] Operations: 1. Event-Scenario Mapping: Construct an event knowledge graph based on financial business logic, define key journey nodes (such as "financial return calculation completed" and "application interrupted"), and associate them with business scenarios (such as "unsatisfactory financial return" and "perception of process complexity"). Example: Event: "Exit financial page" → Scenario Tags: {"Unsatisfactory return", "Terminology comprehension barrier"}. 2. Dynamic Rule Engine: Conditional Matching: Match user operation events with a preset rule base (e.g., novice user + "exit financial page" → triggers the "terminology comprehension barrier" scenario); Real-time Context Enhancement: Integrate behavioral features such as the user's current operation time and page scrolling speed (through embedded data) to calculate the scenario confidence score (e.g., scrolling speed > 2 seconds / screen → "unfocused browsing", reducing the scenario weight).

[0129] Output: Dynamic scenario labels (e.g., "Unmet profit expectations - confidence level 0.8"); a priority-ranked pool of candidate questions (sorted in descending order of scenario confidence level).

[0130] Step 2.2: Questionnaire generation and path optimization based on reinforcement learning.

[0131] Inputs: dynamic scene tags, candidate question pool; real-time user behavior data (such as the current page focus area, number of operation interruptions).

[0132] Operations: 1. Semantic Dynamic Adaptation: Adjust the question wording based on the user's current operational context (e.g., dynamically replace the general question "Are you satisfied with the rate of return?" with "Are you satisfied with the current annualized rate of return of xx%?"); Terminology Alignment: Verify terminology consistency through a financial knowledge base to avoid ambiguity (e.g., contextual adaptation between "fee rate" and "annualized interest rate"). 2. Question Path Optimization: Define a question value function: Design a reward mechanism based on historical response rates and information gain (e.g., the decrease in entropy after a user answers); Dynamically balance exploration (pushing new questions) and utilization (prioritizing high-value questions), and update the question push strategy. Branch Logic Reinforcement Learning: Train the decision model to adjust subsequent question branches based on the user's real-time answers (e.g., "yes" / "no") (e.g., answer "yes" → jump to "specific reasons for dissatisfaction"; answer "no" → trigger the "suggestion collection" branch).

[0133] Output: Personalized questionnaire link; real-time updated question strategy model parameters.

[0134] Step 2.3: Feedback-driven questionnaire self-optimization mechanism.

[0135] Inputs: User response data (structured ratings, open-ended text); questionnaire response rate, average completion time, and abandonment rate.

[0136] Operations: 1. Feedback Feature Extraction: Text Sentiment-Operation Correlation: Analyze the correlation between user feedback and operation logs using a dual-channel neural network (text channel and behavior channel) (e.g., the strong correlation between "cumbersome steps" and "identity verification steps > 5"); Question Validity Evaluation: Calculate question information entropy and discriminative power (e.g., the contribution of high-discriminative questions to user classification). 2. Incremental Model Training: Online Learning: Update model parameters hourly to adapt to changes in behavioral patterns; Cold Start Optimization: Use meta-learning to quickly adapt the initial strategy for new users.

[0137] Output: Optimized questionnaire template library (stored according to scenario); version iteration record of dynamic question push strategy.

[0138] Example and Effect: Scenario: User D (a novice user) interrupts the "Loan Application" process. 1. Event Awareness: The system detects the "Application Interruption" event, and combined with the user tag (novice) and operation log (process time > 3 minutes), triggers the "Process Complexity Awareness" scenario (confidence level 0.85). 2. Dynamic Generation: Select a high-value question: "Did you abandon your application due to too many material upload steps?" (historical response rate 72%); after the user answers "yes", jump to the subsequent question: "Which of the following steps would you like to simplify first? (Material upload / identity verification / proof of income)".

[0139] 3. Self-optimization: The dual-channel model analysis feedback found a strong correlation between "material upload" and "repeated upload times > 3 times" in the operation log, and generated an optimization suggestion: "Merge material upload into a single operation (priority 1)".

[0140] Process 3: Prediction of the production effect of questionnaire analysis (production effect prediction subsystem).

[0141] Objective: Accurately predict market performance: Based on user survey data and behavior logs, predict user acceptance, market share, and key financial scenario indicators (such as payment conversion rate and wealth management repurchase rate) after product launch.

[0142] Step 3.1: Data preprocessing.

[0143] Inputs: User survey data, structured data: user ratings (1-5 points), feature preferences (multiple selections), sensitivity (range selection); open text data: user feedback (e.g., "cumbersome transfer steps", "difficult-to-understand financial terminology"); system logs: user operation paths (e.g., "number of payment interruptions", "duration spent on the risk assessment page"), error messages (e.g., identity verification failure types); external data: industry benchmark indicators (e.g., average payment success rate).

[0144] Operations: 1. Data cleaning and validity filtering: Remove invalid samples (response time <10 seconds, duplicate submissions, logically contradictory answers); handle outliers. 2. Feature engineering and structured feature extraction: User acceptance index: Weighted calculation of scores and function usage frequency (e.g., weekly transfer usage); financial behavior indicators, payment success rate, wealth management page bounce rate, risk assessment interruption rate. Text feature extraction: Extract text keywords (e.g., "cumbersome steps" → "process complexity", "high risk" → "security concerns"); construct sentiment polarity labels (positive / negative / neutral). 3. Data balancing: If the data distribution is skewed (e.g., high satisfaction rate >70%), generate minority class samples and remove noise.

[0145] Output: Cleaned structured dataset (user ID × 150-dimensional features); text feature matrix (user ID × keyword weights); balanced training and test sets (7:3 ratio).

[0146] Step 3.2: Prediction model selection and training.

[0147] Input: Preprocessed dataset (structured + text features).

[0148] Operation: 1. Acceptance prediction model: Model training: Use user acceptance index as the dependent variable and function usage frequency, payment success rate, and text sentiment polarity as independent variables; Feature importance analysis: Quantify the contribution of each feature to the prediction results (e.g., "number of payment steps" contributes 35%); Classification prediction extension: For users with low acceptance (predicted value < 60 points), classify them into specific types (e.g., "process dissatisfaction type" and "performance dissatisfaction type").

[0149] 2. Performance Prediction Model: Time Series Prediction: Predicts the payment conversion rate trend for the next 3 months, taking into account seasonal factors; Ensemble Learning Optimization: Integrates user profiles and industry data to output the market share probability distribution.

[0150] 3. Improvement suggestion generation: Text analysis: Identify high-frequency issues in negative feedback (such as "slow identity verification"); Rule mapping: Preset financial scenario optimization rule base (e.g., if the mention rate of "cumbersome steps" is >20%, then trigger "process simplification suggestion"); Prioritization: Generate an improvement list based on the degree of impact of the problem and business objectives.

[0151] Output: User acceptance prediction report (including continuous scores and category tags); market share performance probability distribution chart (e.g., 95% confidence interval for payment conversion rate); list of improvement suggestions (e.g., optimize identity verification to within 3 steps, priority 1).

[0152] Process 4: Multi-dimensional attribution analysis of functional defects (attribution analysis and suggestion generation subsystem).

[0153] Objective: To accurately pinpoint the functional aspects that users are dissatisfied with, provide actionable optimization suggestions, and achieve automated attribution by correlating structured user feedback with operation logs.

[0154] Step 4.1: Correlation modeling between user feedback and operation logs.

[0155] 1. Feedback Data Preprocessing: Text Segmentation and Semantic Analysis: Open-ended text feedback is segmented to extract key entities (e.g., "multiple identity verification steps" and "delayed investment returns"), and semantic role labeling identifies the specific objects of user pain points (e.g., "multiple steps" → "identity verification process"). Sentiment Polarity Mapping: Keywords are assigned sentiment scores (e.g., "cumbersome" = -0.8, "efficient" = +0.7), generating sentiment feature vectors.

[0156] 2. Structural Processing of Operation Logs: Log Event Labeling: Convert events in the operation log (such as "Authentication Steps = 5" and "Investment Return Calculation Time = 8 seconds") into structured tags. Process Branch Coverage Statistics: Analyze the branch coverage rate of user operation paths (such as "50% of users give up in step 3") and quantify the frequency of abnormal events (such as "Authentication Failure Rate = 15%)".

[0157] 3. Dynamic Association Matching: Rule Engine and Similarity Calculation: Construct a mapping rule base between feedback keywords and log tags (e.g., "multiple steps" → "number of identity verification steps"), and combine it with the cosine similarity algorithm to match association relationships that are not covered by rules. Time Sequence Alignment: Incorporate operation log data within the user feedback time window (±5 minutes) into the association analysis to ensure causal time sequence consistency.

[0158] Step 4.2: Attribution analysis of multi-model collaboration.

[0159] 1. Model Input and Data Format:

[0160] Feature matrix: Integrates user feedback features (keyword weights, sentiment polarity) and operation log features (number of steps, time taken, error rate) to form a user identifier × multimodal feature matrix (example: [User C: {"identification steps":5,"multiple steps":0.9, "time taken":8 seconds}]).

[0161] 2. Collaboration Mechanism:

[0162] First-level attribution: Using user satisfaction as the dependent variable, the model is trained to identify global key factors (such as the contribution of "number of identity verification steps" accounting for 45%), and outputs a ranking of feature importance.

[0163] Secondary attribution: Based on the key features of the output, refine the attribution path (e.g., "Number of steps > 3 → Satisfaction decreases by 20%), and pinpoint the specific defective link (e.g., "Step 3 takes too long").

[0164] 3. Model structure improvement:

[0165] Dual-channel neural network: Parallel input channels are designed to process text features and log features separately, and dynamic weighted fusion is achieved through an attention mechanism to improve association accuracy. Reinforcement learning optimization: Model hyperparameters (such as learning rate and tree depth) are adjusted, with attribution accuracy as the reward function, to achieve adaptive tuning.

[0166] Step 4.3: Interpretability output and suggestion generation.

[0167] Generate a feature contribution heatmap (e.g., "number of identity verification steps" accounts for 62% of the negative feedback).

[0168] Priority rule engine: Generates a list of optimization suggestions based on attribution results and business goals (e.g., suggesting "reduce the authentication steps from 5 to 3").

[0169] This embodiment has at least the following effects.

[0170] 1. Precise User Segmentation and Dynamic Scenario Adaptation. Through financial scenario behavior path analysis (such as operational efficiency and risk sensitivity) and clustering, fine-grained user segmentation (e.g., novice users / expert users) is achieved, solving the coarse-grained problem of traditional segmentation methods. Combining real-time user journey events (such as "exiting the wealth management page" or "application interruption") with reinforcement learning, context-sensitive questionnaire questions are dynamically generated, improving response rates and avoiding the static limitations of traditional questionnaires.

[0171] 2. Multidimensional Data Fusion and Automated Attribution Analysis. Through dual-channel feature fusion technology (text analysis + structured processing of operation logs), user feedback (e.g., "cumbersome steps") is accurately correlated with system logs (number of steps, time consumed, error rate), solving the problem of disconnect between feedback and behavioral data in traditional methods. Multi-level attribution is achieved: the first-level model identifies key global factors (e.g., "number of identity verification steps" contributes 45%), and the second-level model refines the defect path (e.g., "step 3 takes too long"), improving attribution accuracy.

[0172] 3. Adaptive optimization and interpretability of the predictive model. Dynamically adjust the questionnaire delivery strategy to balance exploration and utilization, improving data collection efficiency. Generate actionable optimization suggestions (such as "simplify identity verification to within 3 steps, priority 1"), making the direction of technical improvements transparent and verifiable.

[0173] 4. Closed-loop iteration and business value enhancement. A complete closed-loop system, from user behavior analysis to dynamic questionnaire generation, effect prediction, and attribution optimization, supports real-time data-driven product strategy iteration. By accurately predicting deployment results and attributing defects, resource allocation efficiency is improved, ineffective development investment is reduced, and the effectiveness of product strategies is validated.

[0174] Corresponding to the above method embodiments, this application also provides an apparatus embodiment. Figure 3 The diagram illustrates a structural block diagram of a user feedback information analysis device according to an embodiment of this application.

[0175] like Figure 3 As shown, the user feedback information analysis device 300 in this embodiment includes: a correspondence unit 310, an association unit 320, and a demand unit 330.

[0176] The correspondence unit 310 is used to determine the correspondence between user feedback information and user operation logs; any user feedback information and its corresponding user operation log are determined for the same user; the user operation log is used to represent a sequence of operation information. In one embodiment, the correspondence unit 310 can be used to execute the operation S210 described above, which will not be repeated here.

[0177] The association unit 320 is configured to, for any user feedback information, determine the operation information associated with the user feedback information from the sequence of operation information represented by the corresponding user operation log, and determine the user's improvement requirements based on the user feedback information and the determined operation information. In one embodiment, the association unit 320 may be used to perform the operation S220 described above, which will not be repeated here.

[0178] The requirement unit 330 is used to determine the overall improvement requirements based on user improvement requirements identified from different user feedback information. In one embodiment, the requirement unit 330 can be used to perform the operation S230 described above, which will not be repeated here.

[0179] Optionally, the correspondence unit 310 is used to: obtain the user's authorization to use the user operation log to analyze user feedback information; and, if the user's authorization to use the user operation log to analyze user feedback information is obtained, determine the correspondence between the user feedback information and the user operation log.

[0180] Optionally, the association unit 320 is used to: input the target user feedback information and the corresponding user operation log into a pre-trained analysis model for any user feedback information, and determine the operation information associated with the target user feedback information predicted by the analysis model from the operation information sequence represented by the corresponding user operation log; the analysis model is used to extract features from the input user feedback information and user operation log through different feature extraction modules respectively.

[0181] Optionally, the association unit 320 is used to: determine user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction based on the user feedback information and the determined operation information; the requirement unit 330 is used to: determine overall improvement needs based on the user improvement needs determined for different user feedback information and the degree of impact of the user improvement needs on user satisfaction.

[0182] Optionally, the device further includes: an emotion unit, configured to: determine user emotion information in response to any user feedback information; an association unit 320, configured to: determine user improvement needs based on the user feedback information, the determined operation information, and the determined user emotion information; and a demand unit 330, configured to: determine overall improvement needs based on the user improvement needs determined for different user feedback information and the user emotion information determined for different user feedback information.

[0183] Optionally, the device further includes: an emotion unit, used to: determine user emotion information in response to any user feedback information; and an association unit 320 used to: determine user improvement needs and the degree of impact of the determined user improvement needs on user satisfaction based on the user feedback information, the determined operation information, and the determined user emotion information.

[0184] Optionally, the requirement unit 330 is used to: determine the requirement priority of user improvement requirements based on user improvement requirements identified for different user feedback information; and determine user improvement requirements with a requirement priority greater than a preset priority threshold as overall improvement requirements.

[0185] Optionally, the method for determining user feedback information includes: for any business process, pushing a preset set of questions to the user after the business process has ended; determining user feedback information based on the user's response to the preset set of questions; the preset set of questions is determined based on the business process in question.

[0186] For an explanation of the device embodiment, please refer to the explanation of other embodiments.

[0187] According to embodiments of this application, any multiple modules among the correspondence unit 310, association unit 320, demand unit 330, and emotion unit can be merged into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the correspondence unit 310, association unit 320, demand unit 330, and emotion unit can be at least partially implemented as hardware circuitry, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any appropriate combination of any of these three implementation methods. Alternatively, at least one of the correspondence unit 310, association unit 320, demand unit 330, and emotion unit can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0188] Figure 4 A block diagram schematically illustrates an electronic device suitable for implementing a user feedback information analysis method according to an embodiment of this application.

[0189] like Figure 4As shown, an electronic device 900 according to an embodiment of this application includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0190] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 902 and / or RAM 903. It should be noted that the programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0191] According to embodiments of this application, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0192] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0193] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 902 and / or RAM 903 and / or one or more memories other than ROM 902 and RAM 903 described above.

[0194] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement any of the method embodiments provided in the embodiments of this application.

[0195] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0196] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 909, and / or installed from a removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0197] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0198] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0199] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0200] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for analyzing user feedback information, characterized in that, include: Determine the correspondence between user feedback information and user operation logs; Each user feedback message and its corresponding user operation log are determined for the same user; The user operation log is used to represent the sequence of operation information; For any user feedback information, determine the operation information associated with the user feedback information from the operation information sequence represented by the corresponding user operation log, and determine the user's improvement needs based on the user feedback information and the determined operation information. Based on the user improvement needs identified from different user feedback, the overall improvement needs are determined.

2. The method according to claim 1, characterized in that, Determining the correspondence between user feedback information and user operation logs includes: Obtain user authorization to use user action logs for analyzing user feedback. Once the user has authorized the use of user operation logs to analyze user feedback information, the correspondence between user feedback information and user operation logs is determined.

3. The method according to claim 1, characterized in that, The step of determining the operation information associated with any user feedback information from the operation information sequence represented by the corresponding user operation log includes: For any user feedback information, the user feedback information and the corresponding user operation log are input into a pre-trained analysis model to determine the operation information associated with the user feedback information predicted by the analysis model from the operation information sequence represented by the corresponding user operation log. The analysis model is used to extract features from the input user feedback information and user operation logs through different feature extraction modules.

4. The method according to claim 1, characterized in that, The step of determining user improvement needs based on the user feedback information and the determined operation information includes: determining user improvement needs based on the user feedback information and the determined operation information, and the degree of impact of the determined user improvement needs on user satisfaction. The process of determining overall improvement requirements based on user improvement needs identified from different user feedback information includes: determining overall improvement requirements based on user improvement needs identified from different user feedback information and the degree of impact of user improvement needs on user satisfaction.

5. The method according to claim 1, characterized in that, The method further includes: determining user emotion information based on any user feedback information; The step of determining user improvement needs based on the user feedback information and the determined operation information includes: determining user improvement needs based on the user feedback information, the determined operation information, and the determined user emotional information; The process of determining overall improvement requirements based on user improvement needs identified from different user feedback information includes: determining overall improvement requirements based on user improvement needs identified from different user feedback information and user sentiment information identified from different user feedback information.

6. The method according to claim 4, characterized in that, The method further includes: determining user emotion information based on any user feedback information; The process of determining user improvement needs based on the user feedback information and the determined operational information, and the degree of impact of the determined user improvement needs on user satisfaction, includes: determining user improvement needs based on the user feedback information, the determined operational information, and the determined user emotional information, and the degree of impact of the determined user improvement needs on user satisfaction.

7. The method according to claim 1, characterized in that, The determination of overall improvement requirements based on user improvement needs identified from different user feedback information includes: Based on the user improvement needs identified from different user feedback information, the priority of user improvement needs is determined; user improvement needs with a priority greater than a preset priority threshold are identified as overall improvement needs.

8. The method according to claim 1, characterized in that, The methods for determining the user feedback information include: For any given business process, once the business process has ended, a set of pre-defined questions will be pushed to the user. Based on the user's responses to the preset set of questions, determine the user feedback information; The preset set of questions is determined based on the business process being targeted.

9. A user feedback information analysis device, characterized in that, include: The correspondence unit is used to determine the correspondence between user feedback information and user operation logs; Each user feedback message and its corresponding user operation log are determined for the same user; the user operation log is used to represent the sequence of operation information. The association unit is used to determine the operation information associated with the user feedback information from the operation information sequence represented by the corresponding user operation log, and to determine the user's improvement needs based on the user feedback information and the determined operation information. The requirement unit is used to determine the overall improvement requirements based on user improvement requirements identified from different user feedback information.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

Images