Credit report generation method and device

Abstract

One or more embodiments of the invention provide a credit report generation method and device, and the method comprises the steps: firstly obtaining historical behavior data of a target user on a credit investigation platform, and determining a target user portrait of the target user based on the historical behavior data corresponding to the target user; then acquiring historical behavior data of other users matched with the portrait of the target user on the credit investigation platform, and processing the historical behavior data corresponding to the target user and the historical behavior data corresponding to other users based on a prediction model obtained by training to generate a predicted credit label of the target user; the predicted credit label is used for representing a predicted credit attention direction of a user group to which the target user belongs; and finally, acquiring historical credit investigation data of the target user, and generating a credit report which takes the predicted credit label as a content generation guide for the target user based on the predicted credit label and the historical credit investigation data.

Description

Technical Field

[0001] This specification relates to one or more embodiments in the field of credit report generation technology, and in particular to a credit report generation method and apparatus. Background Technology

[0002] With the deepening of the construction of the social credit system, personal credit reports provided by credit reporting agencies have become an important basis for assessing credit status.

[0003] However, the credit reports in related technologies lack the ability to proactively perceive user interests and preferences, resulting in homogenized credit report information. This leads to a large amount of redundant data in the credit reports, requiring users to manually filter key information, which easily causes technical problems such as high usage barriers and low information retrieval efficiency. Summary of the Invention

[0004] In view of this, one or more embodiments of this specification provide a method and apparatus for generating credit reports.

[0005] To achieve the above objectives, one or more embodiments of this specification provide the following technical solutions: According to a first aspect of one or more embodiments of this specification, a method for generating a credit report is provided, comprising: Obtain historical behavioral data of the target user on the credit reporting platform, and determine the target user profile of the target user based on the historical behavioral data corresponding to the target user; The system acquires historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs; Obtain the target user's credit history data, and based on the predicted credit tag and the credit history data, generate a credit report for the target user with the predicted credit tag as the content generation guide.

[0006] According to a second aspect of one or more embodiments of this specification, a credit report generation apparatus is provided, comprising: The acquisition module acquires the target user's historical behavior data on the credit reporting platform and determines the target user profile based on the historical behavior data corresponding to the target user. The prediction module acquires historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs; The generation module acquires the target user's historical credit data and, based on the predicted credit tag and the target user's historical credit data, generates a credit report for the target user with the predicted credit tag as the content generation guide.

[0007] According to a third aspect of one or more embodiments of this specification, an electronic device is provided, comprising: processor; Memory used to store processor-executable instructions; The processor implements the method as described in the first aspect by running the executable instructions.

[0008] According to a fourth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method as described in the first aspect.

[0009] According to a fifth aspect of one or more embodiments of this specification, a computer program product is provided, comprising: a computer program / instructions that, when executed by a processor, implement the method as described in the first aspect.

[0010] As can be seen from the above embodiments, the credit report generation method and apparatus provided in one or more embodiments of this specification first acquires the historical behavior data of a target user on a credit reporting platform, and determines the target user profile of the target user based on the historical behavior data corresponding to the target user; then, it acquires the historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on a trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs; finally, it acquires the target user's historical credit data, and generates a credit report for the target user with the predicted credit tag as the content generation guide based on the predicted credit tag. When generating a credit report for a target user, a predictive model is first used to predict the target user's credit focus direction to obtain a predicted credit tag. Then, the target user's credit report is generated simultaneously using the predicted credit tag and the target user's credit history data. This ensures that the credit report is generated with the predicted credit tag as the content guide, avoiding the inclusion of a large amount of content that users are not interested in. As a result, users can quickly find the information they are interested in from the credit report, thus improving the service quality of the credit report. Attached Figure Description

[0011] Figure 1 This is an exemplary embodiment of the architecture diagram of an application scenario for a credit report generation method.

[0012] Figure 2 This is a flowchart illustrating a credit report generation method provided in an exemplary embodiment.

[0013] Figure 3 This is a flowchart illustrating another credit report generation method provided in an exemplary embodiment.

[0014] Figure 4 This is a schematic diagram of the structure of a device provided in an exemplary embodiment.

[0015] Figure 5 This is a block diagram of a credit report generation apparatus provided in an exemplary embodiment. Detailed Implementation

[0016] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0017] The organizational information (including but not limited to organizational equipment information, organizational personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) involved in this manual are all information and data authorized by the organization or fully authorized by all parties. Furthermore, the collection, use, and processing of such data must comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding operation portals are provided for the organization to choose to authorize or refuse.

[0018] As described in the background section, the credit report generation and presentation mechanisms in related technologies generally lack the ability to proactively perceive and accurately adapt to users' interests, preferences, and core needs. Current mainstream credit reports mostly use standardized templates for information output. Whether for ordinary individual consumers, high-frequency loan users, or low-frequency credit inquiries, the reports they receive are highly homogenized in terms of content structure, indicator display dimensions, and information priority ranking. This results in credit reports generated by related technologies failing to tailor information filtering to the user's actual usage scenarios (such as mortgage applications, car loan applications, rental fulfillment, and workplace background checks), nor optimizing information presentation logic based on the user's past query habits and focus areas. This homogenized design directly leads to credit reports containing a large amount of redundant data unrelated to the user's current needs. For example, when a user only needs to check recent loan repayment records to apply for a new loan, the report is piled with details of small consumer loans settled years ago, public service payment records unrelated to the credit scenario (non-core indicators), and redundant basic identity information. This redundant data not only consumes additional storage and transmission resources but also imposes a significant information filtering burden on users. This forces users to manually locate and extract key information directly relevant to their needs (such as core credit scores, recent delinquency records, large debts, and key performance records) from lengthy and complex reports. Therefore, in related technologies, users must manually sift through redundant information in credit reports, easily leading to technical problems such as high barriers to entry and low information retrieval efficiency.

[0019] In summary, to address the technical problem of low user query efficiency caused by excessive redundant information in credit reports, this specification provides a credit report generation method. First, it acquires the historical behavioral data of a target user on a credit reporting platform and determines the target user profile based on this historical behavioral data. Since different types of users focus on different modules when using a credit reporting platform, the user's historical behavioral data can accurately reflect their true profile. After obtaining the target user profile, considering that the sample size is small and it's difficult to truly reflect all the target user's areas of interest, it further acquires the historical behavioral data of other users matching the target user profile on the credit reporting platform. This allows the predictive model to predict the target user's areas of interest using more relevant historical behavioral data.

[0020] The trained prediction model, after being inputted with historical behavioral data corresponding to the target user and other users, processes this data to generate a predicted credit tag for the target user. This predicted credit tag characterizes the predicted credit focus of the user group to which the target user belongs. Finally, the model acquires the target user's historical credit data and, based on the predicted credit tag and the historical credit data, generates a credit report for the target user with the predicted credit tag as the content generation guide. Because the predicted credit tag is obtained by first predicting the target user's credit focus using the prediction model, subsequent credit report generation can be based on this predicted credit tag to generate a report that better aligns with the expectations of the user's user group. This avoids the inclusion of a large amount of content that users are not interested in, allowing users to quickly find information of interest within the credit report and improving the service quality of the credit report.

[0021] Figure 1 This is an exemplary embodiment illustrating the architecture of a credit report generation method for a specific application scenario. Figure 1 As shown, the method may include a server 11, a network 12, and several electronic devices, such as a PC (Personal Computer) 13, a mobile phone 14, etc.

[0022] Server 11 can be a physical server containing a single host, or it can be a virtual server hosted in a host cluster. During operation, server 11 can run server-side programs for a specific application to implement its functions. For example, when server 11 runs a credit report generation program, it can implement the entire process of the credit report generation method. That is, server 11 can process the historical behavior data of the target user and the historical behavior data of other users matching the target user profile based on the trained prediction model to generate a predicted credit tag for the target user. Based on the predicted credit tag and the target user's credit history data, it can generate a credit report for the target user with the predicted credit tag as the content generation guide.

[0023] PC13 and mobile phone14 are just some of the types of electronic devices that organizations can use. In reality, organizations can obviously also use electronic devices such as tablets, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smartwatches, etc.), etc., and one or more embodiments in this specification do not limit this. During operation, the electronic device can run a client-side program of an application to implement the relevant functions of that application. For example, when the electronic device runs the program service of the aforementioned credit report generation method, it can act as a client for that program. That is, the user can send a request to the server to generate a credit report through the electronic device and receive the credit report returned by the server. The client application of the aforementioned program service can be launched and run on the electronic device. The client-side program can be a native application installed on the electronic device, or it can be a mini-program, quick app, or other similar form. Of course, when using web technologies such as HTML5 or similar, the relevant functions can be implemented through a page displayed by a browser. Here, the browser can be a standalone browser application or a browser module embedded in some applications.

[0024] As for the network 12 that enables interaction between electronic devices such as PC13 and mobile phone 14 and server 11, wired or wireless networks can be selected for communication based on the communication methods supported by the respective electronic devices. This specification does not impose any restrictions on this. For example, PC13 can support both wired and wireless communication, so it can use either wired or wireless networks as needed. Mobile phone 14 typically only supports wireless communication, so it can use a wireless network for communication.

[0025] refer to Figure 2 Here is a flowchart of a credit report generation method provided in this specification, which includes the following steps: S202, obtain the target user's historical behavior data on the credit reporting platform, and determine the target user profile of the target user based on the historical behavior data corresponding to the target user.

[0026] The target user is the user for whom a credit report needs to be generated. This user can be any user; the use of the target user designation is merely for differentiation from other users, not for limiting the user. To accurately determine the target user profile, historical behavioral data of the target user on credit reporting platforms can be obtained. Since different types of users focus on different aspects when using credit reporting platforms, the corresponding historical behavioral data will also differ. Therefore, by using the target user's historical behavioral data on credit reporting platforms, the target user profile can be accurately determined. It should be noted that the aforementioned credit reporting platforms can be one or more, without limitation.

[0027] It should be noted that the user's historical behavior data on the credit reporting platform is used to characterize various user behaviors on the platform, reflecting the user's level of interest in different credit information or modules. In some embodiments of this specification, the historical behavior data includes the user's log records on the credit reporting platform, the user's interaction records on the credit reporting platform, and the user's credit reporting service usage records on the platform. The credit reporting service usage records store the number of times and the duration of use of various types of credit reporting services on the platform. The user's log records on the credit reporting platform are generally used to record various usage traces on the platform, including page browsing records, data access records, platform login times, etc.

[0028] To more accurately reflect users' interest in different credit reporting modules, this embodiment further obtains user interaction behavior records on the credit reporting platform. These records generally reflect user states during the use of the platform, and since these states do not directly affect data changes on the platform, they are generally not saved in the log. Considering that user interest in certain credit information can influence their own state—for example, if a particular information module on the platform prompts prolonged thought, even without any action taken, the user may still be interested—this specification effectively collects this information through user interaction behavior records (page dwell time) on the credit reporting platform. In some embodiments, these records include user click heatmaps and user page dwell time.

[0029] In some embodiments of this specification, determining the target user profile based on the historical behavior data corresponding to the target user includes: Extract feature data from multiple preset dimensions from the historical behavior data corresponding to the target user; The feature data of multiple preset dimensions corresponding to the target user are input into the trained profile construction model to obtain the target user profile.

[0030] To accurately determine the target user profile, in this embodiment, multiple preset dimensions of feature data are extracted from the target user's historical behavior data on the credit reporting platform. Compared to a single dimension, extracting features from multiple preset dimensions helps the model better understand the user's intent across different dimensions. In some embodiments, the multiple preset dimensions can be set as needed, and there is no limitation thereto. To more accurately depict the user's focus on credit data, in some embodiments, the multiple preset dimensions of feature data include: user data query frequency, user's time sensitivity to credit reports, and the user's attention weight to various modules in the credit reporting platform. In some embodiments, multiple preset dimensions of feature data can be extracted from the target user's historical behavior data using a pre-trained feature extraction model, or historical behavior data can be populated into the features of each preset dimension using a preset rule algorithm, and there is no limitation thereto.

[0031] After extracting feature data from multiple preset dimensions corresponding to the target user, this feature data can be input into a trained profile building model to obtain the target user profile through the output of the model. In some embodiments, multiple fields of the user profile can be predefined, and then the profile building model can determine the field values ​​of each field from the feature data of multiple preset dimensions, ultimately forming the user profile from the field values. It should be noted that the above profile building model can be a traditional neural network module or a large language model, and there is no limitation on this.

[0032] In some embodiments, after predefining multiple fields of a user profile, a default value can be set for each field. Then, an initial user profile is generated based on the default values ​​of each field. Finally, the field values ​​of each field in the initial user profile are updated using feature data of multiple preset dimensions to obtain the user profile.

[0033] S204, obtain historical behavior data of other users matching the target user profile on the credit reporting platform, and process the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs.

[0034] After defining the user profile of the target user, it's important to consider that, as an individual, the target user's historical behavioral data on the credit reporting platform is limited. Furthermore, when a specific group is generally interested in an event, as a member of that group, one is likely to maintain interest in that event as well. However, this interest may be affected by factors such as time constraints or information transmission discrepancies, and cannot be immediately reflected in an individual's historical behavioral data. For example, User A belongs to a group interested in used cars. If a loan incentive policy for purchasing a used car is recently released, User A, as a member of this group, is likely to be interested in the relevant credit information. However, in reality, User A may receive the information about the loan incentive policy late, or User A may be too busy to log in to the credit reporting platform, resulting in their historical behavioral data not being updated. In this case, relying solely on User A's individual historical behavioral data may not accurately predict the level of interest of User A's user group in the aforementioned information. However, if the historical behavioral data of multiple users within a group interested in used cars is used to predict the focus of user A's user group, it is highly likely to reflect user A's true level of interest in the aforementioned used car-related information. Therefore, in the embodiments of this specification, when predicting the focus of a target user, the historical behavioral data of other users matching the target user profile on the credit reporting platform are first obtained. Then, based on a trained prediction model, the historical behavioral data corresponding to the target user and the historical behavioral data corresponding to the other users are processed to generate a predicted credit tag for the target user. This predicted credit tag is used to characterize the predicted credit focus of the user group to which the target user belongs.

[0035] It should be noted that the number of other users matching the target user profile can be one or more, and there is no limitation on this. Generally, the more other users there are, the more accurate the corresponding prediction result will be. In some embodiments, when obtaining the historical behavior data of other users matching the target user profile on the credit reporting platform, other users whose user profiles are the same as or similar to the target user profile can be found through the target user profile, and then the historical behavior data of those other users on the credit reporting platform can be obtained.

[0036] To further improve the efficiency of acquiring historical behavior data of other users matching the target user profile on the credit reporting platform, in some embodiments of this specification, the credit report generation method further includes: Other users who match the target user profile are saved to the target user's associated user set.

[0037] To improve the speed of identifying users with the same user profile, after identifying other users matching the target user profile, these other users can be saved to the target user's associated user set. This way, when the target user needs to access the historical behavior data of other users matching the target user profile on the credit reporting platform, they can directly find these other users through the target user's associated user set. Simultaneously, when another user also needs to generate a credit report, multiple users who may match that user profile can be found at once through the target user's associated user set.

[0038] To improve the efficiency of obtaining historical behavior data of other users matching the target user profile on the credit reporting platform, in some embodiments of this specification, the other users are multiple; obtaining historical behavior data corresponding to other users matching the target user profile includes: Obtain any other user that matches the target user profile; In response to determining that any other user is associated with a set of associated users, historical behavior data corresponding to each user in the set of associated users of any other user is obtained, as historical behavior data corresponding to other users that match the target user profile; wherein, all users in the set of associated users match the same user profile.

[0039] To improve the efficiency of acquiring other users matching the target user profile, when there are multiple other users matching the target user profile, upon acquiring any other user matching the target user profile, it can be first determined whether that other user is associated with a set of associated users. If it is determined that the other user is associated with a set of associated users, the historical behavior data corresponding to each user in the set of associated users is acquired, and used as the historical behavior data corresponding to other users matching the target user profile; wherein, all users in the set of associated users match the same user profile. It should be noted that a set of associated users can be generated for each user in advance, or multiple users with the same user profile can be saved into the same set of associated users each time a credit report is generated.

[0040] To accurately determine the predicted credit tag of a target user, in some embodiments of this specification, the historical behavior data corresponding to the target user and the historical behavior data corresponding to other users are processed based on a trained prediction model to generate the predicted credit tag of the target user, including: For each user among the target user and the other users, feature data of multiple preset dimensions are extracted from the historical behavior data corresponding to each user. The feature data of each user corresponding to multiple preset dimensions are input into the trained prediction model to generate the predicted credit tag of the target user.

[0041] To accurately generate the predicted credit tag for the target user, in this embodiment, multiple preset-dimensional feature data are extracted from the historical behavior data of each user, including the target user and other users. Compared to a single dimension, extracting features from multiple preset dimensions helps the model more accurately predict the target user's focus on credit information, thus accurately obtaining the target user's predicted credit tag. It should be noted that the prediction model can be a traditional neural network model or a large language model. The specific training process of the prediction model can be found in related technologies, and will not be elaborated upon here. To more accurately help the model understand the user's focus on credit data, in some embodiments of this specification, the multiple preset-dimensional feature data include: user data query frequency, user time sensitivity to credit reports, and user attention weights to various modules in the credit reporting platform.

[0042] S206, obtain the target user's credit history data, and based on the predicted credit tag and the credit history data, generate a credit report for the target user with the predicted credit tag as the content generation guide.

[0043] After determining the predicted credit tags used to characterize the target user's predicted credit focus, in order to accurately generate a credit report, the target user's credit history data can be obtained first. This credit history data generally refers to some credit data that the user has already generated. This credit history data may include the target user's bank transaction records, repayment records, default records, overdue records, occupational information, and income certificates, etc. Unlike related technologies that directly generate a full credit report based on the target user's credit history data, in the embodiments of this specification, when generating a credit report for the target user, both the target user's credit history data and the target user's predicted credit tags are considered simultaneously to generate a credit report with the predicted credit tags as the content generation guide. Compared with the full credit report generated in related technologies, this credit report with the predicted credit tags as the content generation guide reduces a large amount of redundant information that users are not interested in, making it convenient for users to quickly find the information they need from the credit report.

[0044] It should be noted that the credit report generated for the target user based on the predicted credit tag may be a credit report containing only content related to the predicted credit tag, or it may provide a detailed introduction to the content related to the predicted credit tag while briefly introducing content unrelated to the predicted credit tag. Alternatively, in some embodiments, the layout of the credit report content may be adjusted based on the predicted credit tag, so that the content related to the predicted credit tag is placed at the beginning or in a position that is easier for the user to see.

[0045] In some embodiments of this specification, based on the predicted credit tag and the credit history data, a credit report is generated for the target user with the predicted credit tag as the content generation guide, including: Based on the predicted credit tag, a target credit field matching the predicted credit tag is determined from multiple preset credit fields; Based on the target user's credit history data, determine the field information corresponding to the target credit field; Based on the field information corresponding to the target credit field, a credit report is generated for the target user with the predicted credit tag as the content.

[0046] To accurately generate a credit report guided by the predicted credit tag, a target credit field matching the predicted credit tag can first be determined from multiple preset credit fields. These preset credit fields can be set in advance as needed. In some embodiments, the preset credit fields can be various credit fields from a full credit report that includes as much information as possible; that is, all preset credit fields can form a full credit report covering all information. The target credit field matching the predicted credit tag among the multiple preset credit fields is generally a credit field of interest to the target user. After determining the target credit field matching the predicted credit tag, the field information corresponding to the target credit field can be determined based on the target user's credit history data. Then, a credit report guided by the predicted credit tag can be generated for the target user based on the field information corresponding to the target credit field. In some embodiments, the credit report guided by the predicted credit tag can be directly composed of the field information corresponding to the target credit field. In some embodiments, the credit report guided by the predicted credit tag can also include other information besides the field information corresponding to the target credit field; for distinction, this other information can be generated in a simplified manner.

[0047] In some embodiments, determining the field information corresponding to the target credit field based on the target user's credit history data may include: inputting the target user's credit history data and the target credit field into a large speech module, and having the large language model output the field information corresponding to the target credit field. Alternatively, in some embodiments, keyword matching can be used to filter the field information corresponding to the target credit field from the target user's credit history data. It should be noted that the target credit field may be multiple or one, and this is not limited.

[0048] In some embodiments of this specification, based on the predicted credit tag and the target user's credit history data, a credit report is generated for the target user with the predicted credit tag as the content generation guide, including: Based on the predicted credit tags and the target user's credit history data, prompt words are generated and input into a large language model to generate a credit report for the target user with the predicted credit tags as the content guide.

[0049] In the embodiments of this specification, prompt words can also be directly generated based on the predicted credit tag and the target user's credit history data, and then input into a large language model to generate a credit report for the target user with the predicted credit tag as the content generation guide. It should be noted that in some embodiments, the predicted credit tag and the target user's credit history data can be filled into a preset template, and then the prompt words can be generated using the filled preset template. For example, in one example, the prompt word could be: "Please generate a credit report for the target user based on the target user's credit history data XXX and with the predicted credit tag XXXX as the content generation guide."

[0050] In some embodiments of this specification, the credit report generation method further includes: Recommend credit reporting services that match the target user's credit report to the target user.

[0051] To better serve the target users, after generating a credit report for the target users, credit reporting services that match the target users' credit reports can be recommended to the target users, so that the target users can improve some indicators in their credit reports through these credit reporting services.

[0052] In some embodiments of this specification, the credit report generation method further includes: The target user's time sensitivity to information reporting is determined based on the target user's historical behavior data; Based on the time sensitivity, the update frequency for generating credit reports for the target user, guided by the predicted credit tags, is determined.

[0053] Considering that different users have varying degrees of sensitivity to the time sensitivity of credit reports—for example, some users are highly sensitive and will frequently refresh their credit reports to observe whether certain indicators have changed, while others are less sensitive and may refresh their credit reports less frequently, or even if they do refresh, they may not pay much attention to changes in certain indicators—and that generating a new credit report incurs costs for the system, this embodiment of the specification, in order to save costs while meeting the needs of different types of users, determines the target user's time sensitivity to information reports based on their historical behavioral data, and then determines the update frequency of generating credit reports for the target user based on the predicted credit tags. In some embodiments, the time sensitivity is directly proportional to the credit report update frequency; that is, the higher the time sensitivity, the higher the corresponding credit report update frequency.

[0054] In some embodiments, a trained sensitivity prediction model can be used to process historical behavioral data corresponding to a target user to determine the target user's time sensitivity to information reporting. It should be noted that the sensitivity prediction model can be a traditional neural network model or a large language model, and there is no limitation thereto.

[0055] In some embodiments of this specification, the credit report generation method further includes: In response to a credit report request issued by the target user, a target time difference is determined between the current credit report request issued by the target user and the previous credit report request issued by the target user. In response to determining that the target time difference is less than the time difference corresponding to the update frequency, the credit report corresponding to the last credit report request is returned to the target user; In response to determining that the target time difference is greater than or equal to the time difference corresponding to the update frequency, a new credit report with the predicted credit tag as the content generation guide is generated for the target user, and the new credit report generated for the target user is returned to the target user.

[0056] Considering that different users have different sensitivities to credit reports, when a credit report request is received from a target user, the target time difference between the current request and the previous request can be determined first. Furthermore, the relationship between this target time difference and the time difference corresponding to the credit report update frequency can be assessed. If the target time difference is less than the time difference corresponding to the update frequency, the credit report corresponding to the previous request can be returned to the target user. Since the target user's request frequency now exceeds the credit report update frequency corresponding to their time sensitivity, returning the previous request reduces the number of credit report generation times while meeting user needs. If the target time difference is greater than or equal to the time difference corresponding to the update frequency, a new credit report with the predicted credit tags as its content generation guide needs to be generated for the target user, and the newly generated report is returned to the target user. It should be noted that the process of regenerating the credit report for the target user can refer to the process described above of generating a credit report for the target user with the predicted credit tags as the content guide. The difference between each generation of the credit report lies in the time of obtaining historical behavioral data and credit history data.

[0057] refer to Figure 3 This is a flowchart illustrating another credit report generation method provided in the embodiments of this specification. The method includes the following steps: S302, Data Acquisition.

[0058] The collected data generally includes users' historical behavioral data on the credit reporting platform and their historical credit data. Historical behavioral data includes users' log records on the credit reporting platform, records of users' interaction behavior on the platform, and records of users' use of credit reporting services on the platform. The records of users' interaction behavior on the credit reporting platform include: user click heatmaps and user page dwell time.

[0059] S304, Feature Extraction.

[0060] The extracted features generally include the frequency of user data queries, the user's time sensitivity to credit reports, and the weight of the user's attention to each module in the credit reporting platform.

[0061] S306, User Profile Building.

[0062] After extracting multi-dimensional features from a user's historical behavioral data on a credit reporting platform, a user profile can be constructed based on these extracted features. In some embodiments, feature data corresponding to a user in multiple preset dimensions can be input into a trained profile construction model to obtain the user profile.

[0063] S308, focus on direction prediction.

[0064] After obtaining a user profile, the historical behavior data of other users matching the user profile on the credit reporting platform can be obtained first. Based on the trained prediction model, the historical behavior data corresponding to the user and the historical behavior data corresponding to other users are processed to generate the user's predicted credit tag. The predicted credit tag is used to characterize the user's predicted credit focus direction. S310 generates personalized credit reports.

[0065] After obtaining the user's predicted credit tag, a personalized credit report is generated for the user based on the predicted credit tag and the user's credit history data, with the predicted credit tag as the content generation guide.

[0066] This specification provides a credit report generation method. First, it acquires the historical behavioral data of a target user on a credit reporting platform and then determines a target user profile based on this historical behavioral data. Since different types of users focus on different modules when using the credit reporting platform, the user's historical behavioral data can accurately reflect their true profile. After acquiring the target user profile, considering that the sample size for a single user is small and it's difficult to truly reflect all of the target user's true areas of interest, it further acquires the historical behavioral data of other users matching the target user profile on the credit reporting platform. This allows the predictive model to predict the target user's areas of interest using more relevant historical behavioral data. The trained prediction model, after being inputted with historical behavioral data corresponding to the target user and other users, processes the historical behavioral data to generate a predicted credit tag for the target user. This predicted credit tag characterizes the predicted credit focus of the user group to which the target user belongs. Finally, the model acquires the target user's historical credit data and, based on the predicted credit tag and the historical credit data, generates a credit report for the target user with the predicted credit tag as the content generation guide. Because the predicted credit tag is obtained by first predicting the target user's credit focus using the prediction model, subsequent credit report generation can be based on this predicted credit tag to generate a credit report that better aligns with the expectations of the user's user group. This avoids the inclusion of a large amount of content that users are not interested in, allowing users to quickly find information of interest within the credit report and improving the service quality of the credit report.

[0067] Figure 4 This is a schematic structural diagram of a device provided in an exemplary embodiment. For example... Figure 4As shown, device 400 mainly consists of a communication interface 402, a mechanism interface 404, a processor 406, and a data storage 408. These components are interconnected and communicate with each other via a method bus, network, or other connection mechanism 410. Communication interface 402 enables device 400 to communicate with other devices, access networks, and transmission networks via analog or digital modulation. For example, communication interface 402 may include a chipset and antenna for wireless communication with a radio access network or access point. Furthermore, communication interface 402 can be a wired interface such as Ethernet, Token Ring, or USB port, or a wireless interface such as Wi-Fi, Bluetooth, Global Positioning System (GPS), or wide area wireless interface (e.g., WiMAX or LTE). Of course, communication interface 402 can also support other forms of physical layer interfaces and standard or proprietary communication protocols. Communication interface 402 may also include multiple physical communication interfaces, such as Wi-Fi, Bluetooth, and wide area wireless interfaces.

[0068] Mechanism interface 404 includes receiving mechanism input and providing output to the mechanism. Therefore, mechanism interface 404 may include input components such as a keypad, keyboard, touch-sensitive or presence-sensitive panel, computer mouse, trackball, joystick, microphone, still camera, and video camera, and output components such as a display screen (which may be combined with a touch-sensitive panel), CRT, LCD, LED, display using DLP technology, printer, and other similar devices known or developed in the future. Mechanism interface 404 may also generate auditory output via speakers, speaker jacks, audio output ports, audio output devices, headphones, and other similar devices known or developed in the future. In some embodiments, mechanism interface 404 may include software, circuitry, or other forms of logic capable of transmitting and receiving data from external mechanism input / output devices. Additionally or alternatively, device 400 may support remote access from other devices via communication interface 402 or another physical interface (not shown). Mechanism interface 404 may be configured to receive mechanism input, the position and movement of which may be indicated by an indicator or cursor described herein. Mechanism interface 404 may also be configured as a display device for rendering or displaying text fragments.

[0069] Processor 406 may contain one or more general-purpose processors and / or special-purpose processors.

[0070] Data storage 408 may include one or more volatile and / or non-volatile storage components and may be integrated wholly or partially with processor 406. Data storage 408 may include removable and non-removable components.

[0071] Processor 406 is capable of executing program instructions 418 (e.g., compiled or uncompiled program logic and / or machine code) stored in data storage 408 to perform the various functions described herein. Data storage 408 may comprise a non-transitory computer-readable medium on which program instructions are stored, which, when executed by device 400, enable device 400 to perform any methods, processes, or functions disclosed in this specification and / or the accompanying drawings. Processor 406 executing program instructions 418 may result in processor 406 using data 412.

[0072] For example, program instructions 418 may include an operation method 422 (e.g., an operation method kernel, device driver, and / or other modules) installed on device 400 and one or more applications 420 (e.g., a browser, social application, or game application). Similarly, data 412 may include operation method data 416 and application data 414. Operation method data 416 is primarily accessible to operation method 422, while application data 414 is primarily accessible to one or more applications 420. Application data 414 may reside in file methods visible or hidden from the device 400.

[0073] Application 420 can communicate with operation method 422 through one or more application programming interfaces (APIs). These APIs help application 420 read and / or write application data 414, transmit or receive information via communication interface 402, receive or display information on mechanism interface 404, etc.

[0074] In some terminology, application 420 may be simply referred to as "app". Furthermore, application 420 can be downloaded to device 400 through one or more online app stores or app markets. However, applications can also be installed on device 400 in other ways, such as through a web browser or a physical interface on device 400 (e.g., a USB port).

[0075] Please refer to Figure 5 Credit report generation devices can be applied to, for example... Figure 4 The apparatus shown is used to implement the technical solution described in this specification. The credit report generation device may include: The acquisition module 502 acquires the historical behavior data of the target user on the credit reporting platform, and determines the target user profile of the target user based on the historical behavior data corresponding to the target user. The prediction module 504 acquires historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs; The generation module 506 acquires the target user's historical credit data and, based on the predicted credit tag and the target user's historical credit data, generates a credit report for the target user with the predicted credit tag as the content generation guide.

[0076] In some embodiments of this specification, the acquisition module is specifically used for: Extract feature data from multiple preset dimensions from the historical behavior data corresponding to the target user; The feature data of multiple preset dimensions corresponding to the target user are input into the trained profile construction model to obtain the target user profile.

[0077] In some embodiments of this specification, the prediction module is specifically used for: For each user among the target user and the other users, feature data of multiple preset dimensions are extracted from the historical behavior data corresponding to each user. The feature data of each user corresponding to multiple preset dimensions are input into the trained prediction model to generate the predicted credit tag of the target user.

[0078] In some embodiments of this specification, the feature data of the multiple preset dimensions include: user data query frequency, user time sensitivity to credit reports, and user attention weight to each module in the credit reporting platform.

[0079] In some embodiments of this specification, the generation module is specifically used for: Based on the predicted credit tag, a target credit field matching the predicted credit tag is determined from multiple preset credit fields; Based on the target user's credit history data, determine the field information corresponding to the target credit field; Based on the field information corresponding to the target credit field, a credit report is generated for the target user with the predicted credit tag as the content.

[0080] In some embodiments of this specification, the generation module is specifically used for: Based on the predicted credit tags and the target user's credit history data, prompt words are generated and input into a large language model to generate a credit report for the target user with the predicted credit tags as the content guide.

[0081] In some embodiments of this specification, the credit report generation apparatus further includes a service recommendation module, used for: Recommend credit reporting services that match the target user's credit report to the target user.

[0082] In some embodiments of this specification, the credit report generation apparatus further includes a frequency module for: The target user's time sensitivity to information reporting is determined based on the target user's historical behavior data; Based on the time sensitivity, the update frequency for generating credit reports for the target user, guided by the predicted credit tags, is determined.

[0083] In some embodiments of this specification, the credit report generation apparatus further includes a credit report return module, used for: In response to a credit report request issued by the target user, a target time difference is determined between the current credit report request issued by the target user and the previous credit report request issued by the target user. In response to determining that the target time difference is less than the time difference corresponding to the update frequency, the credit report corresponding to the last credit report request is returned to the target user; In response to determining that the target time difference is greater than or equal to the time difference corresponding to the update frequency, a new credit report with the predicted credit tag as the content generation guide is generated for the target user, and the new credit report generated for the target user is returned to the target user.

[0084] In some embodiments of this specification, the historical behavior data includes the user's log records on the credit reporting platform, the user's interaction records on the credit reporting platform, and the user's credit reporting service usage records on the credit reporting platform.

[0085] In some embodiments of this specification, the user's interaction behavior records on the credit reporting platform include: user click heatmaps and user page dwell time.

[0086] In some embodiments of this specification, the credit report generation apparatus further includes an association module for: Other users who match the target user profile are saved to the target user's associated user set.

[0087] In some embodiments of this specification, the other users are multiple; the acquisition module is specifically used for: Obtain any other user that matches the target user profile; In response to determining that any other user is associated with a set of associated users, historical behavior data corresponding to each user in the set of associated users of any other user is obtained, as historical behavior data corresponding to other users that match the target user profile; wherein, all users in the set of associated users match the same user profile.

[0088] For ease of description, the above devices are described by dividing them into various modules or units based on their functions. Of course, when implementing one or more of these specifications, the functions of each module or unit can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another method, or some features may be ignored or not executed.

[0089] Based on the same concept as the above method, this specification also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor executes the executable instructions to implement the steps of the credit report generation method as described in any of the above embodiments.

[0090] Based on the same concept as the methods described above, this specification also provides a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the steps of the credit report generation method as described in any of the above embodiments.

[0091] Based on the same concept as the methods described above, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the credit report generation method as described in any of the above embodiments.

[0092] What those skilled in the art will understand is: In this specification, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded.

[0093] In this specification, “a,” “an,” and “the” do not specifically refer to the singular, but may also include the plural.

[0094] In this specification, ordinal numbers such as "first," "second," etc., do not necessarily indicate order; they are often used to distinguish between objects. For example, "first server" and "second server" usually refer to two servers. To differentiate between these two servers, they are described as "first server" and "second server." Of course, sometimes these two servers may be the same server.

[0095] In this specification, unless explicitly stated otherwise, "receiving and sending data" does not necessarily mean direct receiving and sending; it can also mean indirect receiving and sending. For example, A receiving data sent by B can be understood as A directly receiving the data sent by B, or it can be understood as A indirectly receiving the data sent by B through other entities such as C. Similarly, B sending data to A can be understood as B sending the data directly to A, or it can be understood as B indirectly sending the data to A through other entities such as C. Here, C can be one entity, or it can be two or more entities.

[0096] In this specification, unless explicitly stated otherwise, the relationships between structures can be direct or indirect. For example, when describing "A is connected to B," unless it is explicitly stated that A and B are directly connected, it should be understood that A can be directly connected to B or indirectly connected to B. Similarly, when describing "A is on top of B," unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B or indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0097] This specification uses specific terms to describe embodiments thereof. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of those different embodiments or examples, without contradiction.

[0098] Although one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is only one of many possible execution orders and does not represent the only execution order. Therefore, when the claims involve method steps, any changes or adjustments to the order of such steps, or the parallelism between steps, are also within the scope of protection of the claims.

Claims

1. A method for generating a credit report, comprising: Obtain historical behavioral data of the target user on the credit reporting platform, and determine the target user profile of the target user based on the historical behavioral data corresponding to the target user; The system acquires historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user; the predicted credit tag is used to characterize the predicted credit focus direction of the user group to which the target user belongs; Obtain the target user's credit history data, and based on the predicted credit tag and the credit history data, generate a credit report for the target user with the predicted credit tag as the content generation guide.

2. The method according to claim 1, wherein determining the target user profile of the target user based on the historical behavior data corresponding to the target user includes: Extract feature data from multiple preset dimensions from the historical behavior data corresponding to the target user; The feature data of multiple preset dimensions corresponding to the target user are input into the trained profile construction model to obtain the target user profile.

3. The method according to claim 1, wherein the historical behavior data corresponding to the target user and the historical behavior data corresponding to other users are processed based on the trained prediction model to generate a predicted credit tag for the target user, comprising: For each user among the target user and the other users, feature data of multiple preset dimensions are extracted from the historical behavior data corresponding to each user. The feature data of each user corresponding to multiple preset dimensions are input into the trained prediction model to generate the predicted credit tag of the target user.

4. The method according to claim 2 or 3, wherein the feature data of the plurality of preset dimensions includes: The weighting of user data query frequency, user time sensitivity to credit reports, and user attention to various modules in the credit reporting platform.

5. The method according to claim 1, based on the predicted credit tag and the credit history data, generating a credit report for the target user with the predicted credit tag as the content generation guide, comprising: Based on the predicted credit tag, a target credit field matching the predicted credit tag is determined from multiple preset credit fields; Based on the target user's credit history data, determine the field information corresponding to the target credit field; Based on the field information corresponding to the target credit field, a credit report is generated for the target user with the predicted credit tag as the content.

6. The method according to claim 1, based on the predicted credit tag and the target user's credit history data, generating a credit report for the target user with the predicted credit tag as the content generation guide, comprising: Based on the predicted credit tags and the target user's credit history data, prompt words are generated and input into a large language model to generate a credit report for the target user with the predicted credit tags as the content guide.

7. The method according to claim 1, further comprising: Recommend credit reporting services that match the target user's credit report to the target user.

8. The method according to claim 1, further comprising: The target user's time sensitivity to information reporting is determined based on the target user's historical behavior data; Based on the time sensitivity, the update frequency for generating credit reports for the target user, guided by the predicted credit tags, is determined.

9. The method according to claim 8, further comprising: In response to a credit report request issued by the target user, a target time difference is determined between the current credit report request issued by the target user and the previous credit report request issued by the target user. In response to determining that the target time difference is less than the time difference corresponding to the update frequency, the credit report corresponding to the last credit report request is returned to the target user; In response to determining that the target time difference is greater than or equal to the time difference corresponding to the update frequency, a new credit report with the predicted credit tag as the content generation guide is generated for the target user, and the new credit report generated for the target user is returned to the target user.

10. The method according to claim 1, wherein the historical behavior data includes the user's log records on the credit reporting platform, the user's interaction behavior records on the credit reporting platform, and the user's credit reporting service usage records on the credit reporting platform.

11. The method according to claim 10, wherein the user's interaction behavior records on the credit reporting platform include: User click heatmap and user page dwell time.

12. The method according to claim 1, further comprising: Other users who match the target user profile are saved to the target user's associated user set.

13. The method according to claim 1, wherein the other users are multiple; obtaining historical behavior data corresponding to other users matching the target user profile includes: Obtain any other user that matches the target user profile; In response to determining that any other user is associated with a set of associated users, historical behavior data corresponding to each user in the set of associated users of any other user is obtained, as historical behavior data corresponding to other users that match the target user profile; wherein, all users in the set of associated users match the same user profile.

14. A credit report generation apparatus, comprising: The acquisition module acquires the target user's historical behavior data on the credit reporting platform and determines the target user profile based on the historical behavior data corresponding to the target user. The prediction module acquires historical behavior data of other users matching the target user profile on the credit reporting platform, and processes the historical behavior data corresponding to the target user and the historical behavior data corresponding to the other users based on the trained prediction model to generate a predicted credit tag for the target user. The predicted credit tag is used to characterize the predicted credit focus of the user group to which the target user belongs; The generation module acquires the target user's historical credit data and, based on the predicted credit tag and the target user's historical credit data, generates a credit report for the target user with the predicted credit tag as the content generation guide.

15. An electronic device comprising: processor; Memory used to store processor-executable instructions; The processor executes the executable instructions to implement the method as described in any one of claims 1-13.

16. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1-13.

17. A computer program product comprising: A computer program / instruction that, when executed by a processor, implements the method as described in any one of claims 1-13.

Images