Data processing method and device, equipment, storage medium and program product

Abstract

This application provides a data processing method, apparatus, device, storage medium, and program product, relating to fintech and other fields. The method includes: generating a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information, wherein the multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. When a target user logs into a target website, advertising information corresponding to the target multi-dimensional user features is pushed to the user based on the target browser fingerprint corresponding to the target user, wherein the target multi-dimensional user features are the multi-dimensional user features corresponding to the target user. This method improves the accuracy of advertising information push.

Description

Technical Field

[0001] This application relates to financial technology and other fields, and in particular to a data processing method, apparatus, device, storage medium and program product. Background Technology

[0002] In the financial sector, with the accelerated pace of digitalization, financial institutions are shifting their service models from traditional offline channels to a three-dimensional ecosystem integrating online and offline channels. Corporate online banking portals, as the core digital entry point connecting banks and corporate clients, are no longer limited to basic account management and transaction operations; they are gradually evolving into key marketing platforms for financial institutions to showcase products and services, convey brand value, and reach target customer groups. At the same time, corporate user needs are becoming increasingly diversified and personalized, with significant differences in the financial product needs of companies of different sizes, industries, and stages of development. However, current advertising on bank portals struggles to accurately meet the needs of corporate users, resulting in poor advertising accuracy.

[0003] Therefore, improving the accuracy of advertising information delivery on bank portal websites is an urgent problem to be solved. Summary of the Invention

[0004] This application provides a data processing method, apparatus, device, storage medium, and program product to solve the technical problem of poor accuracy in advertising push on bank portal websites.

[0005] Firstly, this application provides a data processing method, including:

[0006] A user's browser fingerprint is generated based on multi-dimensional user features collected from the user's historical login information. The multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features.

[0007] When a target user enters the target website, advertising information corresponding to the target multidimensional user characteristics is pushed to the user based on the target browser fingerprint corresponding to the target user. The target multidimensional user characteristics are the multidimensional user characteristics corresponding to the target user.

[0008] Optionally, generating the user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information includes:

[0009] Obtain hardware information, software information, browser information, and enterprise information of the devices used by users during their historical logins;

[0010] The multidimensional user features are generated based on the hardware information, the software information, the browser information, and the enterprise information.

[0011] Based on the similarity matching results of the multi-dimensional user features corresponding to the multiple historical logins of each user, a browser fingerprint of each user is generated.

[0012] Optionally, generating the multi-dimensional user features based on the hardware information, the software information, the browser information, and the enterprise information includes:

[0013] The user identity characteristics are determined based on the hardware information and the software information;

[0014] The behavioral preference characteristics are determined based on the browser information;

[0015] The enterprise identity characteristics are determined based on the enterprise information;

[0016] The multidimensional user features are generated based on the user identity features, the behavioral preference features, and the enterprise identity features.

[0017] Optionally, generating browser fingerprints for each user based on the similarity matching results of multi-dimensional user features corresponding to multiple historical logins of each user includes:

[0018] The user type of each user is determined based on the similarity of the user identity features and the first similarity threshold, the similarity of the behavioral preference features and the second similarity threshold, and the similarity of the enterprise identity features and the third similarity threshold corresponding to the multiple historical logins of each user.

[0019] Based on the user type and multidimensional user characteristics corresponding to each user, a browser fingerprint is generated for each user.

[0020] Optionally, generating the browser fingerprint for each user based on the user type and the multidimensional user features includes:

[0021] The initial browser fingerprint is generated by encrypting the multi-dimensional user features corresponding to each user.

[0022] The browser fingerprint of each user is generated by combining the initial browser fingerprint and the user type.

[0023] Optionally, determining the user type of each user based on the similarity of the user identity features corresponding to multiple historical logins of each user and a first similarity threshold, the similarity of the behavioral preference features and a second similarity threshold, and the similarity of the enterprise identity features and a third similarity threshold includes:

[0024] If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of the enterprise identity features is greater than or equal to the third similarity threshold, then the user is determined to be a first user type.

[0025] If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of the enterprise identity features is less than the third similarity threshold, then the user is determined to be a second user type.

[0026] If the similarity of the user identity features is less than the first similarity threshold, the similarity of the behavioral preference features is less than the second similarity threshold, and the similarity of the enterprise identity features is greater than or equal to the third similarity threshold, the user is determined to be a third user type.

[0027] If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the enterprise identity features do not exist, the user is determined to be a fourth user type.

[0028] If the similarity of the user identity features is less than the first similarity threshold, the similarity of the behavioral preference features is less than the second similarity threshold, and the similarity of the enterprise identity features is less than the third similarity threshold, then the user is determined to be a fifth user type.

[0029] Optional, also includes:

[0030] Based on the browser fingerprint corresponding to each user, obtain advertising information corresponding to the multi-dimensional user features corresponding to each user;

[0031] The advertising information corresponding to the multi-dimensional user characteristics of each user will be sent to the target website.

[0032] The step of pushing advertising information corresponding to the target multidimensional user characteristics to the target user based on the target browser fingerprint corresponding to the target user when the target user enters the target website includes:

[0033] Based on the target browser fingerprint, determine the target user type corresponding to the target multidimensional user features;

[0034] Push advertising information corresponding to the target user type to the target user.

[0035] Optional, also includes:

[0036] The browser fingerprint is stored on both the server and the user's browser.

[0037] Secondly, this application provides a data processing apparatus, comprising:

[0038] The processing module is used to generate a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. The multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features.

[0039] The control module is used to push advertising information corresponding to the target multidimensional user features to the target user when the target user logs into the target website, based on the target browser fingerprint corresponding to the target user. The target multidimensional user features are the multidimensional user features corresponding to the target user.

[0040] Thirdly, this application provides an electronic device, including: a processor and a memory; the processor and the memory are communicatively connected.

[0041] The memory stores computer-executed instructions;

[0042] The processor executes computer execution instructions stored in the memory to implement the method as described in any one of the first aspects.

[0043] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the data processing method as described in any one of the first aspects.

[0044] Fifthly, this application provides a computer program product, which, when executed by a processor, is used to implement the data processing method as described in any one of the first aspects.

[0045] The data processing method, apparatus, equipment, storage medium, and program products provided in this application generate a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. These multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. When a target user enters a target website, advertising information corresponding to the target multi-dimensional user features is pushed to the user based on the target user's corresponding browser fingerprint. These multi-dimensional user features are the multi-dimensional user characteristics corresponding to the target user. Using the multi-dimensional dynamic features integrated from the browser fingerprint as a basis, this achieves accurate differentiation of customer needs and personalized matching of advertising content, reducing the problem of homogeneous advertising in traditional push notifications, thereby improving the accuracy of advertising information push on bank portal websites. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0047] Figure 1 A schematic diagram illustrating an advertising placement scenario provided for this application;

[0048] Figure 2 A flowchart illustrating a data processing method provided in an embodiment of this application;

[0049] Figure 3 A flowchart illustrating another data processing method provided in an embodiment of this application;

[0050] Figure 4 A flowchart illustrating another data processing method provided in an embodiment of this application;

[0051] Figure 5 A flowchart illustrating another data processing method provided in an embodiment of this application;

[0052] Figure 6 This is a schematic diagram of the structure of a data processing system provided in an embodiment of this application;

[0053] Figure 7 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application;

[0054] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0055] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0056] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0057] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application 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 the relevant data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, have taken necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize or refuse.

[0058] Furthermore, the technical solution involved in this application, which involves big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.) and the use of artificial intelligence technology for automated decision-making, and makes decisions that have a significant impact on personal rights based on the results of automated decision-making, provides users with corresponding operation entry points for users to choose to agree to or reject the results of automated decision-making; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0059] It should be noted that the data processing methods, apparatus, devices, storage media, and program products provided in this application can be used in financial technology and other fields, as well as in any field other than financial technology and other fields. The application fields of the data processing methods, apparatus, devices, storage media, and program products in this application are not limited.

[0060] To facilitate understanding, we will first provide a detailed introduction to the methods currently used in the banking sector for pushing advertising information on bank portal websites.

[0061] Figure 1 This is a schematic diagram illustrating an advertising placement scenario provided in this application. For example... Figure 1 As shown, this scenario includes: the operating system, the bank's server, and the bank's portal website.

[0062] The operations system is used by bank staff to maintain the advertising information to be deployed. This system can be, for example, an internal bank advertising management system or a third-party advertising platform. Advertising information can include, for example, product posters, pop-up ads, text links, and video ads. When maintaining advertising information, different advertising content can be maintained for different advertising channels and different customer groups based on the advertising deployment channel, advertising content, and target audience, thus forming the advertising information.

[0063] The bank's server-side component retrieves and maintains advertising information from the operational system and distributes this information to the bank's portal website. The portal website then receives and displays the advertising information in specific locations within the website. This server-side component can be, for example, the bank's core business system responsible for distributing advertising data, or a separately deployed advertising middleware service. The bank's portal website can be, for example, a corporate online banking portal or an embedded webpage within a corporate mobile banking application.

[0064] In this way, users can view advertising information on the bank's website after entering the website.

[0065] However, existing methods for pushing advertising information on bank portals have the following problems:

[0066] 1. If a user is not logged in when accessing the bank's portal website, it is impossible to obtain the user's identity characteristics (such as company size, industry attributes, historical transaction behavior, etc.). The system can only rely on static advertising content or simple channel differentiation to display advertising information, making it impossible to achieve personalized advertising based on dynamic customer segmentation. Therefore, the flexibility and accuracy of advertising information delivery are poor.

[0067] 2. In the existing process, all users see completely identical advertising content, which makes it difficult for advertising positions on the bank's portal website to achieve precise targeting value, resulting in problems such as wasted bank operating resources and limited business development.

[0068] In view of this, this application provides a data processing method that generates a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. These multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. When a target user enters a target website, advertising information corresponding to the target multi-dimensional user features is pushed to the user based on the target user's corresponding browser fingerprint. These multi-dimensional user features are the multi-dimensional user characteristics corresponding to the target user. Using the multi-dimensional dynamic features integrated from the browser fingerprint as a benchmark, this method achieves accurate differentiation of customer needs and personalized matching of advertising content, reducing the problem of homogeneous advertising in traditional push notifications, thereby improving the accuracy of advertising information push on bank portal websites.

[0069] The following describes the technical solution of this application and how it solves the aforementioned technical problems through specific embodiments, using the server as the execution entity. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0070] Figure 2This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 2 As shown, the method may include:

[0071] S201. Generate the user's browser fingerprint based on the multi-dimensional user features collected from the user's historical login information.

[0072] Among them, multidimensional user characteristics include user identity characteristics, enterprise identity characteristics, and behavioral preference characteristics.

[0073] In a real-world network environment, each time a user logs into a website, a series of login history records are generated. This information contains a wealth of data, reflecting user characteristics from various perspectives. User identity characteristics can be based on basic information entered during account registration, such as account type and registered email address; this information serves as the user's fundamental identifier in the online world. Enterprise identity characteristics can be obtained through information related to the enterprise associated with the user, such as the enterprise's business area and market position. Behavioral preference characteristics are derived by analyzing the user's specific actions on the website, such as page browsing order and the number of times specific content is clicked.

[0074] To generate a user's browser fingerprint, the collected multidimensional user features can be preprocessed. For different types of features, standardization is necessary because their data formats and units may differ. For example, age and registration duration information in user identity features can be mapped to the same numerical range, such as between 0 and 1, to ensure comparability between features.

[0075] Next, feature selection is performed on the standardized multidimensional user features. This step aims to filter out the features that are most representative and discriminative for generating browser fingerprints. Feature importance assessment methods, such as feature importance ranking based on correlation analysis or machine learning algorithms, can be used to remove features that are not very useful for distinguishing users. The filtered features are then encoded. For categorical features, such as the business domain in enterprise identity features, one-hot encoding can be used to convert them into binary vectors, with each vector representing a business domain category. For numerical features, their standardized values ​​can be directly retained.

[0076] Finally, the encoded features are combined in a specific order to form a fixed-length vector, which is the user's browser fingerprint. This fingerprint uniquely identifies the user and plays a crucial role in subsequent applications such as ad delivery.

[0077] For example, taking a user's access to a corporate online banking portal as an example, the process of collecting the user's historical login information could be as follows: When a user accesses the corporate online banking portal and inserts their USB key to log in, a series of information about the user is collected. This information could include, for example, the hardware information, software information, browser information, and corporate information of the device used by the user during previous logins. This information is collected every time the user accesses the corporate online banking portal and is used as the user's historical login information. Subsequently, multi-dimensional user characteristics of each user can be obtained based on this historical login information, and the user's browser fingerprint can be further generated.

[0078] Subsequently, the user's browser fingerprint can be stored either on the user's browser or on both the user's browser and the bank's server. This way, when the user re-enters the corporate online banking portal, if the browser fingerprint has not been cleared, the user's identity can be directly verified based on the stored fingerprint; if the browser fingerprint has been cleared, the corresponding browser fingerprint can be retrieved from the bank's server.

[0079] S202. When a target user enters the target website, based on the target browser fingerprint corresponding to the target user, push advertising information corresponding to the target multidimensional user characteristics to the user.

[0080] Among them, the target multidimensional user features are the multidimensional user features corresponding to the target user.

[0081] When a target user re-enters the corporate online banking portal, the target browser fingerprint of that user can be obtained. If the browser fingerprint has not been cleared on the user's browser, the target browser fingerprint of that user can be obtained directly from the user's browser. If the browser fingerprint has been cleared on the user's browser, the target browser fingerprint corresponding to the user's browser can be obtained from the bank's server.

[0082] Since the target browser fingerprint is generated based on the target user's multi-dimensional user characteristics, advertising information corresponding to these characteristics can be determined based on the target browser fingerprint. For example, matching advertising information can be filtered from an advertising database based on the target browser fingerprint, and these filtered ads can be pushed to the target user. This could be done by displaying ads in specific locations on a website page or through pop-up windows.

[0083] The method provided in this application generates a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. These multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. When a target user enters a target website, advertising information corresponding to the target multi-dimensional user features is pushed to the user based on the target user's corresponding browser fingerprint. These multi-dimensional user features are the multi-dimensional user characteristics corresponding to the target user. Using the multi-dimensional dynamic features integrated from the browser fingerprint as a basis, the method achieves accurate differentiation of customer needs and personalized matching of advertising content, reducing the problem of homogeneous advertising in traditional push notifications, thereby improving the accuracy of advertising information push on bank portal websites.

[0084] The following section provides a detailed explanation of how to generate a user's browser fingerprint based on the multi-dimensional user features collected from the user's historical login information in step S201. Figure 3 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 3 As shown, the aforementioned step S201 may specifically include:

[0085] S301. Obtain the hardware information, software information, browser information, and enterprise information of the devices used by the user during historical logins.

[0086] The hardware information of the device may include, for example, the device model, the number of central processing unit (CPU) cores, and graphics card information. The software information may include, for example, the system fonts, screen resolution, the applications installed on the device, the time zone, and the geographical location. This information can be used to determine whether the users who enter the target website are the same users, that is, to determine the user's identity characteristics, which is the user's personal identity.

[0087] Browser information can include browser version, browser plugins, bookmarks, cookies, history, etc. Browser information is mainly used to obtain user behavior preferences, such as the types of websites a user frequently visits and the news they follow.

[0088] Enterprise information may include, for example, the enterprise online banking certificate used by the user when logging in (such as the enterprise online banking certificate ID), the category of the enterprise (such as the enterprise groupID), the brand of the U-shield certificate, etc. Enterprise information is mainly used to obtain the user's enterprise identity characteristics in the bank. For example, it may include the industry, asset level, credit rating, risk tolerance, etc. of the enterprise user, and the age, job level, years of service, risk preference, user rating, etc. of the individual user.

[0089] When retrieving information from a user's historical logins, hardware and software information such as device model, CPU core count, and graphics card details can be obtained using browser plugins or dedicated scripts. These programs can detect the device's specific hardware configuration. Software settings such as system fonts and screen resolution can be obtained through the operating system's interface. Information about installed applications can be obtained by querying the system's application list. Time zone and geographic location information can be determined using network positioning technology. Browser information can be obtained through the browser's built-in interfaces. Company information can be obtained from information provided by the user during account registration, company-related interfaces, and the USB security token used.

[0090] S302. Generate multi-dimensional user characteristics based on hardware information, software information, browser information, and enterprise information.

[0091] In one possible implementation, hardware, software, browser, and enterprise information can be comprehensively analyzed. First, this information is categorized and grouped together, with related information grouped into the same category. Then, features are extracted from each category. For the hardware performance category, the device's performance level can be evaluated by calculating a comprehensive index of hardware parameters, such as considering processor performance and memory capacity, to arrive at a hardware performance score. For the software usage category, the frequency and duration of use for different types of software can be statistically analyzed as software usage features. Next, the extracted features are combined. For example, a weighted concatenation method can be used, assigning appropriate weights to features of different categories based on their importance. For instance, features closely related to user identity can be given higher weights, while some auxiliary features can be given lower weights, ultimately forming multi-dimensional user features.

[0092] In another possible implementation, multi-dimensional user characteristics can be generated based on hardware information, software information, browser information, and enterprise information through the following sub-steps:

[0093] S3021. Determine user identity characteristics based on hardware and software information.

[0094] Hardware and software information can provide important clues for identifying user characteristics. Device models and installed applications within the hardware and software information can reflect a user's identity and usage scenarios. For example, a user using a high-end business laptop with a large number of professional office software programs installed is likely a manager or technical professional in a company.

[0095] To determine user identity characteristics, hardware and software information can be preprocessed first. Hardware information is encoded, converting details such as the brand and model of the hardware device into numerical features. Software information is statistically analyzed to calculate metrics such as the number of installations and usage frequency for different types of software.

[0096] Then, machine learning algorithms are used to analyze the preprocessed information. For example, clustering algorithms can be used to group users with similar hardware and software information into the same category, with each category representing a user identity characteristic. For instance, clustering algorithms can categorize users into identity categories such as corporate financial personnel and corporate managers.

[0097] S3022. Determine behavioral preference characteristics based on browser information.

[0098] Browser information meticulously records a user's online behavior, and in-depth analysis of this information can reveal user behavioral preferences. For example, browsing history can reflect a user's areas of interest. Analyzing the types of websites a user visits, such as news websites, shopping websites, and social networking sites, as well as the frequency and duration of visits on different types of websites, can reveal the user's interests. For instance, if a user frequently visits technology news websites and spends a considerable amount of time on these sites, it indicates a high level of interest in technology-related information. Browser search history can reflect a user's immediate needs and interests. Browser bookmarks and favorites information can reflect content that a user has been following for a long time. Websites that a user has bookmarked are usually those they consider valuable and frequently visit; analyzing this bookmark information can further reveal the user's interests and preferences.

[0099] Specifically, browser information can be quantified. Website types in the browsing history can be categorized and statistically analyzed, calculating the access percentage for each type. Keyword extraction and frequency statistics can be performed on search records. Bookmarks and favorites information can be organized and categorized. Then, these quantified results are combined into a behavioral preference feature vector, where each element represents a specific behavioral preference dimension.

[0100] S3023. Determine the enterprise's identity characteristics based on enterprise information.

[0101] Corporate information encompasses various aspects of a company and can be used to identify its unique characteristics. Information such as the company's name, industry, size, and market position depict the company's image and features from different perspectives.

[0102] For company names, relevant information can be obtained by analyzing factors such as naming characteristics and brand influence. Well-known companies often have names with high recognizability and market recognition, while the names of some emerging companies may reflect their innovative business models or development directions.

[0103] The industry a company belongs to is an important dimension of its identity. Companies in different industries have different business characteristics and market demands. For example, companies in the financial industry focus on information related to risk management and investment strategies; while manufacturing companies focus on aspects such as production processes and supply chain management.

[0104] Company size can be measured by indicators such as number of employees, asset size, and operating revenue. Large companies typically have more sophisticated organizational structures, a wider range of business operations, and greater market influence; smaller companies may be more flexible and focused on specific market segments.

[0105] Market position reflects a company's competitive advantage and market share within an industry. Companies that are industry leaders may have advantages in areas such as technology research and development and brand building; while companies with smaller market shares may need to seek development through differentiated competition.

[0106] This enterprise information can be quantified and encoded to form an enterprise identity feature vector. For example, the industry to which the enterprise belongs can be classified and coded, with each industry corresponding to a specific code value.

[0107] S3024. Generate multi-dimensional user characteristics based on user identity features, behavioral preference features, and enterprise identity features.

[0108] After obtaining user identity features, behavioral preference features, and enterprise identity features, they need to be integrated to generate multidimensional user features. For example, a weighted concatenation method can be used to assign appropriate weights to different types of features.

[0109] The user identity feature vector, behavioral preference feature vector, and enterprise identity feature vector are concatenated according to assigned weights to form a multidimensional user feature vector. This multidimensional user feature vector contains multifaceted user characteristics, from individual to occupational, providing a comprehensive and accurate data foundation for subsequent browser fingerprint generation.

[0110] S303. Generate browser fingerprints for each user based on the similarity matching results of the multi-dimensional user features corresponding to each user's multiple historical logins.

[0111] When a user re-enters the corporate online banking portal, relevant information can be collected again and compared with existing information in the database. By comparing the multi-dimensional user features corresponding to each user's multiple historical logins, the similarity matching results of each feature in the multi-dimensional user features are determined. Based on the similarity matching results, the user's identity (including corporate identity and personal identity) is further determined.

[0112] For example, by comparing the multidimensional user characteristics corresponding to multiple historical logins of each user, if the similarity of the users' corporate identity characteristics is high, it indicates that the user's corporate affiliation is stable, and the user is likely an employee of that company. If the similarity of the users' personal identity characteristics is high, it indicates that multiple logins are by the same person. If the similarity of the users' behavioral preference characteristics is high, it can further indicate that multiple logins are by the same person, and so on.

[0113] If the similarity of users' corporate identity characteristics is low, it indicates that the users may serve multiple companies. If the similarity of users' personal identity characteristics is low, it indicates that multiple logins are by different people, possibly a public account of a single company. Similarly, if the similarity of users' behavioral preference characteristics is low, it can further indicate that multiple logins are by different people, and so on.

[0114] The above matching process can be repeated for verification multiple times. The similarity threshold can be determined through trial calculations using a big data model. Multiple verifications, as described above, can improve the accuracy of the comparison results and ensure more reliable judgment of user characteristics.

[0115] Based on this, the similarity matching results of multi-dimensional user features corresponding to each user's multiple historical logins can be used to further determine each user's personal identity, corporate identity, etc., thereby determining each user's user type and generating a browser fingerprint associated with the user type.

[0116] The following section provides a detailed explanation of how, in step S303, browser fingerprints are generated based on the similarity matching results of the multi-dimensional user features corresponding to each user's multiple historical logins. Figure 4 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 4 As shown, the aforementioned step S303 may specifically include:

[0117] S401. Determine the user type of each user based on the similarity of user identity features and the first similarity threshold, the similarity of behavioral preference features and the second similarity threshold, and the similarity of enterprise identity features and the third similarity threshold corresponding to each user's multiple historical logins.

[0118] For each user's multiple historical logins, the similarity between them is calculated using similarity calculation methods. Methods such as cosine similarity or Euclidean distance can be used to compare the user identity feature vectors for each login.

[0119] The calculated user identity feature similarity is compared with a first similarity threshold. If the similarity is greater than or equal to the first similarity threshold, it means that these user identity features are relatively similar; if the similarity is less than the first similarity threshold, it means that these user identity features are significantly different.

[0120] For behavioral preference features, the similarity calculation method is also used to calculate the similarity between behavioral preference features corresponding to multiple historical logins. Then, it is compared with a second similarity threshold. If the similarity is greater than or equal to the second similarity threshold, it indicates that the behavioral preference features are similar; if it is less than the second similarity threshold, it indicates that the behavioral preference features are significantly different.

[0121] For enterprise identity features, the same method is used to calculate similarity and compare it with a third similarity threshold.

[0122] Finally, based on the comparison results corresponding to the above features, the user type of each user can be determined. For example, different combinations of results can correspond to different user types.

[0123] For example, the correspondence between user type and comparison result can be shown in the following example:

[0124] If the similarity of user identity features is greater than or equal to the first similarity threshold, the similarity of behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of enterprise identity features is greater than or equal to the third similarity threshold, the user is identified as the first user type. Specifically, this type of user may be a core employee in a fixed position within an enterprise, such as a long-term financial submitter, finance manager, human resources submitter, or human resources manager. Their device and browser habits are relatively stable, they consistently work for the same enterprise, and their job content and needs are relatively fixed.

[0125] If the similarity of user identity features is greater than or equal to the first similarity threshold, the similarity of behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of enterprise identity features is less than the third similarity threshold, the user is identified as a second user type. In the above scenario, this type of user could be, for example, a professional in financial accounting or bookkeeping services. They use their own fixed devices and browsers, have relatively stable operating habits, but provide services to different companies, so their enterprise identity features will differ. By further analyzing their behavioral preferences in browser information, such as the time periods and operational process preferences when accessing different companies' online banking, as well as multi-dimensional information such as IP address and geographical location in software and hardware information, it is possible to more accurately determine whether it is an individual bookkeeping agent or an accounting agency. For example, if the IP address is relatively dispersed and accesses different companies' online banking at different times, it may be an individual bookkeeping agent serving different companies in different locations; if the IP address is concentrated in a certain office area and has regular access times, it is more likely that an employee of an accounting agency is operating from a unified office location.

[0126] If the similarity of user identity features is less than the first similarity threshold, the similarity of behavioral preference features is less than the second similarity threshold, and the similarity of enterprise identity features is greater than or equal to the third similarity threshold, the user is identified as a third user type. In corporate online banking, this type of user may come from companies with relatively loose management structures or those in their early stages of development. Because these companies are in a developmental phase, staff turnover is high, and there may not be fixed financial personnel or clear job responsibilities. This leads to different individuals using different devices and browsers to log in to corporate online banking, but their enterprise identity remains relatively fixed. For example, an early-stage startup might have multiple employees try handling financial-related tasks. These employees may have different personal identities and operating habits, but they all serve the same company.

[0127] If the similarity of user identity features is greater than or equal to the first similarity threshold, the similarity of behavioral preference features is greater than or equal to the second similarity threshold, and there are no enterprise identity features, the user is identified as the fourth user type. In this scenario, such users may be tourists who use fixed devices and browsers, have relatively stable browsing habits, but have not established an association with a specific enterprise and may only be temporarily accessing the enterprise's online banking portal for general information inquiries.

[0128] If the similarity of user identity features is less than the first similarity threshold, the similarity of behavioral preference features is less than the second similarity threshold, and the similarity of enterprise identity features is less than the third similarity threshold, the user is identified as the fifth user type. In this scenario, this type of user could be a new user. A new user might be logging into the enterprise's online banking portal for the first time and has not yet developed stable characteristics.

[0129] S402. Generate browser fingerprints for each user based on their corresponding user type and multi-dimensional user characteristics.

[0130] One possible implementation is to directly generate each user's browser fingerprint based on their corresponding user type and multidimensional user features. First, the user type is encoded, converting it into a numerical feature. For example, a unique encoding value is assigned to each user type, such as encoding the first user type as 1, the second user type as 2, and so on. Then, the encoded user type features are concatenated with the multidimensional user features. For instance, they can be combined in a specific order to generate each user's browser fingerprint. For example, the user type encoding can be placed before the feature vector, followed by the user identity feature vector, behavioral preference feature vector, and enterprise identity feature vector in sequence.

[0131] Another possible implementation is to encrypt the process of generating each user's browser fingerprint to protect user privacy. Specifically, this can be achieved through the following sub-steps:

[0132] S4021. After encrypting the multi-dimensional user features corresponding to each user, generate the initial browser fingerprint.

[0133] To protect user privacy, multi-dimensional user characteristics need to be encrypted. For example, symmetric or asymmetric encryption algorithms can be used.

[0134] Symmetric encryption algorithms, such as AES, use the same key for both encryption and decryption. When using a symmetric encryption algorithm, an encryption key is first generated, and then this key is used to encrypt a multi-dimensional user feature vector, converting it into an initial browser fingerprint in ciphertext form.

[0135] Asymmetric encryption algorithms use public and private keys for encryption and decryption, such as the RSA algorithm. When generating the initial browser fingerprint, the public key is used to encrypt multi-dimensional user characteristics. The public key can be made public and used to encrypt data; the private key is securely stored by the bank and used to decrypt data. This ensures that even if data is intercepted during transmission, attackers cannot decrypt it to obtain the user's true characteristic information.

[0136] S4022: Combine the initial browser fingerprint and user type to generate browser fingerprints for each user.

[0137] The encrypted initial browser fingerprint is then fused with the user type code. This can be done by concatenating the user type code as a new dimension to the initial browser fingerprint. For example, if the initial browser fingerprint is a vector of length n and the user type code is a single numerical value, then the fused browser fingerprint will be a vector of length n+1.

[0138] The generated browser fingerprint contains multi-dimensional user characteristics and reflects the user's type, while also being encrypted to protect user privacy. Simultaneously, the generated browser fingerprint is stored on both the bank's server and the user's browser, providing dual protection. Storing it on the bank's server ensures data security and integrity, facilitating subsequent comparison and analysis; storing it on the user's browser allows for quick retrieval of fingerprint information upon the user's next login, improving identification efficiency. Furthermore, even if the user clears some data on their browser, the fingerprint information remains on the bank's server, ensuring uninterrupted user identification.

[0139] Figure 5 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 5 As shown, the method may further include:

[0140] S501. Based on the browser fingerprint of each user, obtain advertising information corresponding to the multi-dimensional user characteristics of each user.

[0141] After storing a user's browser fingerprint, a mapping relationship is established between the browser fingerprint and multi-dimensional user characteristics. When it is necessary to push advertisements to a user, the browser fingerprint is used to determine the advertisement information to be pushed.

[0142] Since browser fingerprints are generated based on multi-dimensional user characteristics, the advertising information corresponding to different browser fingerprints is essentially the advertising information corresponding to different multi-dimensional user characteristics. Different advertising matching strategies can be applied to the multi-dimensional user characteristics of different user types.

[0143] For example, for the first user type (such as core personnel in fixed corporate positions mentioned above), if their corporate identity characteristics indicate that the company is in the financial industry and is relatively large, and their behavioral preferences show an interest in financial investment information, then advertisements suitable for high-end investment products and risk management services for large financial companies can be filtered from the advertising database. For the second user type (such as accounting and bookkeeping personnel mentioned above), if their behavioral preferences show an interest in updates and promotional information for accounting software, then advertisements provided by accounting software vendors can be matched, such as introductions to the latest version of accounting software and promotional activities.

[0144] S502. Send the advertising information corresponding to the multi-dimensional user characteristics of each user to the target website.

[0145] After obtaining the advertising information corresponding to each user, this advertising content is downloaded to the bank's portal website. During transmission, it is crucial to ensure data security and integrity. Encrypted transmission protocols, such as SSL / TLS, can be used to encrypt the advertising information, preventing data tampering or theft during transmission.

[0146] Meanwhile, to improve transmission efficiency, advertising information can be compressed to reduce data volume. For example, compression algorithms can be used to compress advertising text and images, which are then decompressed after the portal website receives the data.

[0147] S503. When a target user enters the target website, determine the target user type corresponding to the target multidimensional user characteristics based on the target browser fingerprint.

[0148] When the target user revisits the bank's portal website, the target user's browser fingerprint is obtained. This can be achieved by reading fingerprint information stored in the user's browser (if the user has not cleared it) or by retrieving stored fingerprint information from the bank's server.

[0149] Then, the target user type is determined based on the obtained target browser fingerprint. For example, the target browser fingerprint carries the user type, or the target user type can be determined based on the mapping relationship between the target browser fingerprint and the user type.

[0150] S504. Push advertising information corresponding to the target user type to the target user.

[0151] Based on the identified target user type, select and push advertisements that correspond to that user type from the advertising information already distributed to the portal website.

[0152] The content of the pushed advertisements varies depending on the target user type. For example, if the target user is a corporate finance manager, information on products or services applicable to their industry can be displayed, such as advertisements for financial management solutions or tax planning services for finance managers in the financial industry. If the target user is an individual engaged in accounting or bookkeeping, information on financial products or marketing activities suitable for them can be displayed, such as low-risk financial products suitable for personal investment or promotional offers for personal versions of accounting software. If the target user is a startup company, digital office solutions offered by the bank can be displayed, such as corporate financial management software packages or online office platform services. If the target user is a visitor, information such as account opening instructions and feature introductions can be displayed to guide visitors through the functions and account opening process of corporate online banking, attracting them to become registered users.

[0153] The method provided in this application obtains advertising information corresponding to the multi-dimensional user characteristics of each user based on their browser fingerprint, and then sends this advertising information to the target website. When a target user enters the target website, the target user type corresponding to the target multi-dimensional user characteristics is determined based on the target browser fingerprint, and advertising information corresponding to the target user type is pushed to the target user. By using the browser fingerprint as a link, accurate identification of user characteristics and personalized advertising push are achieved, thereby improving the accuracy and effectiveness of advertising information push on the bank portal website, enhancing the user experience, and establishing a more accurate and efficient information communication bridge between the bank and users.

[0154] To facilitate understanding, the specific implementation of the above data processing method will be exemplified below using the structure of a possible data processing system. Figure 6 This is a schematic diagram of the structure of a data processing system provided in an embodiment of this application. Figure 6 As shown, the system may include: a client and a server for the bank's portal website.

[0155] Each time a user accesses the bank's portal website through its client application, the client can collect the user's login information, such as the hardware, software, and browser information of the device used for login, as well as the company information. The client then sends this information to the server, which can either directly store it as multi-dimensional user characteristics or analyze it to derive multi-dimensional user features.

[0156] The server can perform multiple comparisons based on a user's historical login information (i.e., the login information collected by the client each time mentioned above) to obtain similarity matching results for each feature. Then, it generates a browser fingerprint for each user based on the similarity matching results. After generating the browser fingerprint, one copy can be stored on the server and another on the client. Furthermore, the server can obtain advertising information corresponding to different user types (or customer groups) from external sources (such as an advertising maintenance system).

[0157] When a target user enters the target website, the client can identify the user type (equivalent to identifying the target user's customer group) by using its own saved browser fingerprint or by obtaining a corresponding browser fingerprint from the server. Then, based on the user type, the client filters out the advertising information obtained from the server that corresponds to the target user's user type and displays it, so that the user can receive advertising information that matches their user type.

[0158] Figure 7 This is a schematic diagram of the structure of a data processing apparatus provided in an embodiment of this application. Figure 7 As shown, the data processing device may include: a processing module 11 and a control module 12.

[0159] Processing module 11 is used to generate a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. The multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features.

[0160] Control module 12 is used to push advertising information corresponding to the target multidimensional user characteristics to the target user based on the target browser fingerprint corresponding to the target user when the target user logs into the target website. The target multidimensional user characteristics are the multidimensional user characteristics corresponding to the target user.

[0161] Optionally, processing module 11 is specifically used to obtain hardware information, software information, browser information, and enterprise information of the devices used by the user during historical logins. Based on the hardware information, software information, browser information, and enterprise information, multi-dimensional user features are generated. Based on the similarity matching results of the multi-dimensional user features corresponding to each user's multiple historical logins, browser fingerprints for each user are generated.

[0162] Optionally, processing module 11 is specifically used to determine user identity characteristics based on hardware and software information; determine behavioral preference characteristics based on browser information; determine enterprise identity characteristics based on enterprise information; and generate multi-dimensional user characteristics based on user identity characteristics, behavioral preference characteristics, and enterprise identity characteristics.

[0163] Optionally, processing module 11 is specifically used to determine the user type of each user based on the similarity of user identity features and a first similarity threshold, the similarity of behavioral preference features and a second similarity threshold, and the similarity of enterprise identity features and a third similarity threshold corresponding to each user's multiple historical logins. Based on the user type and multi-dimensional user features, browser fingerprints are generated for each user.

[0164] Optionally, processing module 11 is specifically used to encrypt the multi-dimensional user features corresponding to each user to generate an initial browser fingerprint. The initial browser fingerprint and user type are then combined to generate the browser fingerprint for each user.

[0165] Optionally, processing module 11 is specifically configured to determine the user as a first user type if the similarity of user identity features is greater than or equal to a first similarity threshold, the similarity of behavioral preference features is greater than or equal to a second similarity threshold, and the similarity of enterprise identity features is greater than or equal to a third similarity threshold. If the similarity of user identity features is greater than or equal to the first similarity threshold, the similarity of behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of enterprise identity features is less than the third similarity threshold, the user is determined as a second user type. If the similarity of user identity features is less than the first similarity threshold, the similarity of behavioral preference features is less than the second similarity threshold, and the similarity of enterprise identity features is greater than or equal to the third similarity threshold, the user is determined as a third user type. If the similarity of user identity features is greater than or equal to the first similarity threshold, the similarity of behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of enterprise identity features is less than the third similarity threshold, the user is determined as a fifth user type.

[0166] Optionally, processing module 11 is further configured to obtain advertising information corresponding to the multi-dimensional user characteristics of each user based on the browser fingerprint of each user. Control module 12 is further configured to send the advertising information corresponding to the multi-dimensional user characteristics of each user to the target website. Specifically, processing module 11 is configured to determine the target user type corresponding to the target multi-dimensional user characteristics based on the target browser fingerprint. Control module 12 is specifically configured to push advertising information corresponding to the target user type to the target user.

[0167] Optionally, the control module 12 is also used to store the browser fingerprint on the server side and the user's browser side.

[0168] The data processing apparatus provided in this application embodiment can execute the data processing method in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0169] Figure 8 This is a schematic diagram of an electronic device provided in an embodiment of this application. The electronic device is used to execute the aforementioned data processing method. Figure 8 As shown, the electronic device 800 may include at least one processor 801, a memory 802, and a communication interface 803.

[0170] The memory 802 is used to store programs. Specifically, the program may include program code, which includes computer operation instructions.

[0171] The memory 802 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0172] The processor 801 is used to execute computer execution instructions stored in the memory 802 to implement the method described in the foregoing method embodiments. The processor 801 may be a CPU, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0173] The processor 801 can communicate and interact with external devices through the communication interface 803. In specific implementations, if the communication interface 803, memory 802, and processor 801 are implemented independently, they can be interconnected via a bus to complete communication. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc., but this does not imply that there is only one bus or one type of bus.

[0174] Optionally, in a specific implementation, if the communication interface 803, memory 802, and processor 801 are integrated on a single chip, then the communication interface 803, memory 802, and processor 801 can communicate through an internal interface.

[0175] This application also provides a computer-readable storage medium, which may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. Specifically, the computer-readable storage medium stores program instructions, which are used in the methods described in the above embodiments.

[0176] This application also provides a program product including executable instructions stored in a readable storage medium. At least one processor of a computing device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the computing device to perform the data processing method described above.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A data processing method, characterized in that, include: A user's browser fingerprint is generated based on multi-dimensional user features collected from the user's historical login information. The multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. When a target user enters the target website, advertising information corresponding to the target multidimensional user characteristics is pushed to the user based on the target browser fingerprint corresponding to the target user. The target multidimensional user characteristics are the multidimensional user characteristics corresponding to the target user.

2. The method according to claim 1, characterized in that, The step of generating a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information includes: Obtain hardware information, software information, browser information, and enterprise information of the devices used by users during their historical logins; The multidimensional user features are generated based on the hardware information, the software information, the browser information, and the enterprise information. Based on the similarity matching results of the multi-dimensional user features corresponding to the multiple historical logins of each user, a browser fingerprint of each user is generated.

3. The method according to claim 2, characterized in that, The step of generating the multi-dimensional user features based on the hardware information, the software information, the browser information, and the enterprise information includes: The user identity characteristics are determined based on the hardware information and the software information; The behavioral preference characteristics are determined based on the browser information; The enterprise identity characteristics are determined based on the enterprise information; The multidimensional user features are generated based on the user identity features, the behavioral preference features, and the enterprise identity features.

4. The method according to claim 2, characterized in that, The step of generating browser fingerprints for each user based on the similarity matching results of multi-dimensional user features corresponding to multiple historical logins of each user includes: The user type of each user is determined based on the similarity of the user identity features and the first similarity threshold, the similarity of the behavioral preference features and the second similarity threshold, and the similarity of the enterprise identity features and the third similarity threshold corresponding to the multiple historical logins of each user. Based on the user type and multidimensional user characteristics corresponding to each user, a browser fingerprint is generated for each user.

5. The method according to claim 4, characterized in that, The step of generating browser fingerprints for each user based on the user type and the multidimensional user features includes: The initial browser fingerprint is generated by encrypting the multi-dimensional user features corresponding to each user. The browser fingerprint of each user is generated by combining the initial browser fingerprint and the user type.

6. The method according to claim 4, characterized in that, The step of determining the user type of each user based on the similarity of the user identity features corresponding to multiple historical logins of each user and a first similarity threshold, the similarity of the behavioral preference features and a second similarity threshold, and the similarity of the enterprise identity features and a third similarity threshold, includes: If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of the enterprise identity features is greater than or equal to the third similarity threshold, then the user is determined to be a first user type. If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the similarity of the enterprise identity features is less than the third similarity threshold, then the user is determined to be a second user type. If the similarity of the user identity features is less than the first similarity threshold, the similarity of the behavioral preference features is less than the second similarity threshold, and the similarity of the enterprise identity features is greater than or equal to the third similarity threshold, the user is determined to be a third user type. If the similarity of the user identity features is greater than or equal to the first similarity threshold, the similarity of the behavioral preference features is greater than or equal to the second similarity threshold, and the enterprise identity features do not exist, the user is determined to be a fourth user type. If the similarity of the user identity features is less than the first similarity threshold, the similarity of the behavioral preference features is less than the second similarity threshold, and the similarity of the enterprise identity features is less than the third similarity threshold, then the user is determined to be a fifth user type.

7. The method according to claim 4, characterized in that, Also includes: Based on the browser fingerprint corresponding to each user, obtain advertising information corresponding to the multi-dimensional user characteristics corresponding to each user; The advertising information corresponding to the multi-dimensional user characteristics of each user will be sent to the target website. The step of pushing advertising information corresponding to the target multidimensional user characteristics to the target user based on the target browser fingerprint corresponding to the target user when the target user enters the target website includes: Based on the target browser fingerprint, determine the target user type corresponding to the target multidimensional user features; The advertising information corresponding to the target user type is pushed to the target user.

8. The method according to any one of claims 1-6, characterized in that, Also includes: The browser fingerprint is stored on both the server and the user's browser.

9. A data processing apparatus, characterized in that, include: The processing module is used to generate a user's browser fingerprint based on multi-dimensional user features collected from the user's historical login information. The multi-dimensional user features include user identity features, enterprise identity features, and behavioral preference features. The control module is used to push advertising information corresponding to the target multidimensional user features to the target user based on the target browser fingerprint corresponding to the target user when the target user logs into the target website. The target multidimensional user features are the multidimensional user features corresponding to the target user.

10. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.

Images