Resource niche dynamic reconstruction method and device, resource niche dynamic display method and device, storage medium and computer equipment

Abstract

According to the resource niche dynamic reconstruction method and device, the resource niche dynamic display method and device, the storage medium and the computer equipment, when the business background system receives the resource niche data acquisition request sent by the terminal equipment, diversified user data of the target user bound with the terminal equipment can be acquired, and the user data can be acquired by combining the activity state of the current platform. And forming multi-dimensional data together. The multi-dimensional data is further converted into a multi-dimensional feature vector, and the user value score rate of the target user can be calculated according to the multi-dimensional feature vector. And then, based on the multi-dimensional data and the user value score rate, a target display strategy most suitable for the target user can be intelligently matched, and target resource niche data is acquired according to the target display strategy and then sent to the terminal equipment, so that the terminal equipment can perform dynamic rendering and display. Through the series of dynamic reconstruction and personalized display processes, the user experience is greatly improved, and the user retention rate and the conversion efficiency are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of page display, and in particular to a resource position dynamic reconstruction and display method and device, a storage medium and a computer device. BACKGROUND

[0002] In the current market environment, most traditional e-commerce applications, such as CPS (i.e., commission promotion service paid according to sales), usually adopt a unified page layout and resource position display strategy in user interface design and resource position display strategy. Although this approach may help to quickly build an application framework and attract users in the early stage, over time, this homogenization display strategy gradually exposes its limitations.

[0003] For example, due to the lack of personalized customization for different user needs, this strategy often cannot accurately match the personalized needs of users, resulting in a decline in user experience. Moreover, users may feel disappointed because they cannot find the goods or services they are really interested in, which in turn leads to a decrease in user stickiness and a shortening of use time, thereby reducing user retention rate and conversion efficiency. SUMMARY

[0004] The present application aims to at least solve one of the above technical defects, in particular, the technical defect that the existing e-commerce application adopts a unified page layout and resource position display method, resulting in a low user retention rate and conversion efficiency.

[0005] The present application provides a resource position dynamic reconstruction method, which is applied to a business background system, comprising:

[0006] In response to a resource position data acquisition request sent by a terminal device, user data of a target user bound to the terminal device and the current platform activity state are collected, and multi-dimensional data is formed;

[0007] After the multi-dimensional data is converted into a multi-dimensional feature vector, the user value score rate of the target user is calculated according to the multi-dimensional feature vector;

[0008] Based on the multi-dimensional data and the user value score rate, a target display strategy is matched, and after target resource position data is obtained according to the target display strategy, it is sent to the terminal device.

[0009] Optionally, the collection of the user data of the target user bound to the terminal device and the current platform activity state, and the formation of the multi-dimensional data, comprises:

[0010] The target user bound to the terminal device is determined;

[0011] quantifying a team size of a team in which the target user is located through a social relationship analysis thread;

[0012] obtaining a user portrait of the target user through a user portrait identification thread;

[0013] evaluating a user value level of the target user through a value evaluation thread;

[0014] determining whether the target user is a new user through a new user determination thread;

[0015] detecting an activity state of a current platform through a big promotion state detection thread;

[0016] taking the activity state of the current platform, the team size, the user portrait, the user value level of the target user, and whether the target user is a new user as multi-dimensional data.

[0017] Optionally, the calculating the user value score rate of the target user according to the multi-dimensional feature vector comprises:

[0018] weighting the multi-dimensional feature vector by using a time decay factor to obtain a multi-dimensional weighted feature vector;

[0019] inputting the multi-dimensional weighted feature vector into a pre-configured neural network model to obtain a user value score rate output by the neural network model.

[0020] Optionally, the weighting the multi-dimensional feature vector by using a time decay factor to obtain a multi-dimensional weighted feature vector comprises:

[0021] dividing the multi-dimensional feature vector into a dynamic behavior area and a static attribute area;

[0022] weighting the dynamic behavior area by using exponential decay to obtain a weighted feature vector;

[0023] combining the weighted feature vector and the static attribute area to obtain a multi-dimensional weighted feature vector.

[0024] Optionally, the matching a target display strategy based on the multi-dimensional data and the user value score rate comprises:

[0025] determining a user type and a preliminary display strategy of the target user according to a score rate interval corresponding to the user value score rate;

[0026] matching the multi-dimensional data with a preset scene rule to obtain a matching result;

[0027] determining a target display strategy according to the user type, the preliminary display strategy, and the matching result.

[0028] Optionally, the determining the user type and the preliminary display strategy of the target user according to the score rate interval corresponding to the user value score rate comprises:

[0029] If the user value score rate is located in the first score rate interval, the user type of the target user is determined as a high-value user, and the preliminary display strategy of the target user is a first combination strategy.

[0030] If the user value score rate is located in the second score rate interval, the user type of the target user is determined as a medium-value user, and the preliminary display strategy of the target user is a second combination strategy.

[0031] If the user value score rate is located in the third score rate interval, the user type of the target user is determined as a low-value user, and the preliminary display strategy of the target user is a default strategy.

[0032] Optionally, the multi-dimensional data comprises an activity state of the current platform, a team size of the target user, a user portrait, and whether the target user is a new user.

[0033] The matching the multi-dimensional data with preset scene rules to obtain a matching result comprises:

[0034] When the activity state of the current platform is a large promotion state, and the team size of the target user exceeds a preset size threshold, the intermediate display strategy of the target user is determined as the first combination strategy.

[0035] Otherwise, when the user portrait of the target user meets a specific label, the intermediate display strategy of the target user is determined as the second combination strategy.

[0036] When the user portrait of the target user does not meet the specific label, but the target user is a new user, the intermediate display strategy of the target user is determined as a third combination strategy.

[0037] Otherwise, the intermediate display strategy of the target user is determined as a default strategy.

[0038] Optionally, the user type comprises a high-value user, a medium-value user, and a low-value user.

[0039] The preliminary display strategy comprises a first combination strategy, a second combination strategy, and a third combination strategy.

[0040] The matching result comprises a first combination strategy, a second combination strategy, a third combination strategy, and a default strategy.

[0041] The determining the target display strategy according to the user type, the preliminary display strategy, and the matching result comprises:

[0042] When the user type is a high-value user, the initial display strategy is the first combination strategy, but the matching result is the default strategy, the second combination strategy is used as the target display strategy.

[0043] When the user type is a medium-value user, the initial display strategy is the second combination strategy, but the matching result is the default strategy, the third combination strategy is used as the target display strategy.

[0044] Otherwise, the matching result will be used as the target display strategy.

[0045] This application also provides a method for dynamically displaying resource bits, the method being applied to a terminal device, including:

[0046] Send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system;

[0047] The front-end page is dynamically rendered based on the target resource location data, and the rendered front-end page is displayed to the target user.

[0048] Optionally, the step of dynamically rendering the front-end page based on the target resource location data includes:

[0049] The main thread is invoked to initialize the infrastructure based on the target resource bit data.

[0050] After the main thread is initialized, the user-specific layout calculation task, network image prefetching task, local cache reading task and device performance detection task are executed in parallel according to the target resource bit data through the parallel asynchronous loading pool, and the page skeleton is constructed through the main thread.

[0051] Preemptive rendering is performed on the page skeleton based on the task status of parallel tasks until all resource slots are loaded.

[0052] Optionally, the step of executing user-specific layout calculation tasks, network image prefetching tasks, local cache reading tasks, and device performance detection tasks in parallel using a parallel asynchronous loading pool based on the target resource bit data includes:

[0053] The user-specific layout calculation task and device performance detection task are executed by a parallel asynchronous loading pool based on the target resource bit data.

[0054] The detection results of the device performance detection task are fed back to the decision engine that executes the user-specific layout calculation task through the parallel asynchronous loading pool;

[0055] After the decision engine returns the layout data, it triggers a local cache read task through the parallel asynchronous loading pool, and compares the resource bit version in the local cache with the resource bit version in the target resource bit data to obtain the comparison result.

[0056] After determining the resources to be downloaded based on the comparison results using the parallel asynchronous loading pool, a network image prefetching task is triggered to download the resources.

[0057] This application also provides a resource bit dynamic reconfiguration device, including:

[0058] The data acquisition module is used to respond to the resource bit data acquisition request sent by the terminal device, collect the user data of the target user bound to the terminal device and the current activity status of the platform, and form multi-dimensional data;

[0059] The value calculation module is used to convert the multi-dimensional data into a multi-dimensional feature vector, and then calculate the user value score rate of the target user based on the multi-dimensional feature vector.

[0060] The dynamic reconstruction module is used to match the target display strategy based on the multi-dimensional data and the user value score, and to obtain the target resource position data according to the target display strategy and send it to the terminal device.

[0061] This application also provides a resource location dynamic display device, including:

[0062] The data receiving module is used to send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system.

[0063] The dynamic display module is used to dynamically render the front-end page based on the target resource location data and display the rendered front-end page to the target user.

[0064] This application also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the resource bit dynamic reconstruction method as described in any of the above embodiments, and / or the steps of the resource bit dynamic display method as described in any of the above embodiments.

[0065] This application also provides a computer device, including: one or more processors, and memory;

[0066] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the resource bit dynamic reconstruction method as described in any of the above embodiments, and / or the steps of the resource bit dynamic display method as described in any of the above embodiments.

[0067] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0068] The resource slot dynamic reconstruction and display method, apparatus, storage medium, and computer equipment provided in this application, when the business backend system receives a resource slot data acquisition request sent by the terminal device, can collect diverse user data of the target user bound to the terminal device, including but not limited to user behavior data and user preference data, and combine it with the current platform activity status to form multi-dimensional data. This multi-dimensional data is further transformed into multi-dimensional feature vectors, and the user value score rate of the target user is calculated based on the multi-dimensional feature vectors. This user value score rate can intuitively reflect the user's potential value and purchase intention, providing a scientific basis for formulating differentiated display strategies. Furthermore, based on these multi-dimensional data and user value score rates, this application can intelligently match the most suitable target display strategy for the target user, and accordingly obtain the target resource slot data and send it to the terminal device for dynamic rendering and display. This series of dynamic reconstruction and personalized display processes not only greatly improves the user experience but also effectively increases user retention and conversion efficiency, bringing new development opportunities to e-commerce applications. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 A flowchart illustrating a method for dynamic reconfiguration of resource bits provided in an embodiment of this application;

[0071] Figure 2 A schematic diagram illustrating the matching target display strategy provided in this application embodiment;

[0072] Figure 3 A flowchart illustrating a method for dynamically displaying resource bits, provided in an embodiment of this application;

[0073] Figure 4 A schematic diagram illustrating the process of dynamically rendering a front-end page, provided in an embodiment of this application.

[0074] Figure 5 This is a schematic diagram of the three-tier unified architecture implemented based on the Flutter framework of this application;

[0075] Figure 6 This is a schematic diagram of a resource bit dynamic reconfiguration device provided in an embodiment of this application;

[0076] Figure 7 This is a schematic diagram of the structure of a resource bit dynamic display device provided in an embodiment of this application;

[0077] Figure 8 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0078] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0079] In the current market environment, most traditional CPS (Cost Per Sale, commission-based referral service) shopping guide apps typically adopt a uniform page layout and resource placement strategy in their user interface design and resource placement tactics. While this approach may initially help to quickly build the application framework and attract users, it suffers from significant differences in users' motivations for using shopping guide apps, as detailed below:

[0080] 1. Commission-oriented users: These users primarily focus on high-commission best-selling products zones, promotional order boards, sales ranking competitions, and earnings dashboards. Their core motivation for using shopping guide apps is to earn commissions by promoting products.

[0081] 2. New user group: To improve the first-time user experience and retention conversion rate, strong discount sections such as 0 yuan purchase and Taobao coupon purchase for 1 cent should be the focus of display content.

[0082] 3. Vertical interest users: For example, foodies pay more attention to the nearby merchants and services located by LBS (Location Based Services), so the priority of the food, drink and entertainment section needs to be increased; students pay more attention to electronic products and technology-related sections; mothers tend to prefer the maternal and infant section, live streaming zone, etc.

[0083] 4. High-value users: These users pay more attention to the big brand zone.

[0084] 5. Regular users: The default layout can be maintained.

[0085] The current homogenized display strategy cannot accurately match the user needs of the different types of users mentioned above, resulting in reduced user stickiness and shorter usage time.

[0086] Based on this, this application proposes the following technical solution, as detailed below:

[0087] In one embodiment, such as Figure 1 As shown, Figure 1 This application provides a flowchart illustrating a method for dynamic resource bit reconstruction according to an embodiment of the present application. The present application provides a method for dynamic resource bit reconstruction, which is applied to a business backend system and may include:

[0088] S110: In response to the resource bit data acquisition request sent by the terminal device, collect the user data of the target user bound to the terminal device and the current activity status of the platform, and form multi-dimensional data.

[0089] In this step, when a target user launches the target application or triggers its relevant interface, the terminal device sends a resource bit data acquisition request to the business backend system. Upon receiving this request, the business backend system immediately initiates the data collection process. This process not only covers diverse user data of the target user bound to the terminal device, such as user behavior patterns, historical purchase records, and user preference settings, but also captures the current activity status of the platform, such as whether a promotional activity is underway, and the specific type and scale of the activity. This data is carefully integrated to form a detailed and multi-dimensional data report, laying a solid foundation for subsequent user value assessment and display strategy development.

[0090] In this application, the target user refers to the object currently served by the terminal device. This object can be any individual or entity using the terminal device to access the services provided by the business backend system. In e-commerce shopping guide applications, the target user typically refers to a user who wants to purchase goods and earn commissions or enjoy discounts through promotional links within the application. By comprehensively understanding and deeply analyzing the target user, the business backend system can provide them with a more personalized and precise service experience.

[0091] The target application in this application refers to the shopping guide app or e-commerce platform used by the user. This app or platform works closely with the business backend system to provide users with a range of services, including product promotion, commission earning, and discounts. Within the target application, users can not only browse various product information and participate in various promotional activities, but also earn corresponding sales commissions by sharing promotional links with friends and family. This process not only enriches the user's shopping experience but also brings more traffic and sales opportunities to the e-commerce platform.

[0092] Within the target application, ad placements serve as crucial windows for showcasing product, promotional, and special offers; their layout and display strategies directly impact the user's shopping experience and purchasing decisions. Traditional ad placement display methods often employ a uniform page layout and display strategy, ignoring the differences between users, resulting in a poor user experience and reduced user engagement. This application, however, dynamically restructures ad placements when the target user launches the target application, achieving personalized customization of the display strategy and significantly improving user experience and shopping efficiency.

[0093] Furthermore, when the business backend system responds to a resource bit data acquisition request sent by a terminal device and collects user data of the target user bound to the terminal device and the current activity status of the platform, it can first retrieve the corresponding target user information from the user database using the user identifier carried in the resource bit data acquisition request. This information includes, but is not limited to, the user's basic personal information, historical transaction records, and browsing behavior data. Simultaneously, the system will also capture the current platform's activity status information in real time, such as the start and end times of the activity, the activity type, and participation conditions, ensuring that the collected data is comprehensive and accurate.

[0094] S120: After converting multi-dimensional data into multi-dimensional feature vectors, calculate the user value score of the target user based on the multi-dimensional feature vectors.

[0095] In this step, after collecting user data of the target user bound to the terminal device and the current activity status of the platform through S110 and forming multi-dimensional data, this application can also convert the multi-dimensional data into a multi-dimensional feature vector, and then calculate the user value score rate of the target user based on the multi-dimensional feature vector. In this way, the target display strategy corresponding to the resource position in the target application can be determined based on the user value score rate.

[0096] Specifically, this application utilizes feature engineering to transform multi-dimensional data into multi-dimensional feature vectors. Feature engineering is a crucial step in machine learning, involving extracting useful features from raw data and transforming these features into a format suitable for model training. In this application, feature engineering may include a series of operations such as data cleaning, data transformation, and feature selection to convert the collected multi-dimensional data into multi-dimensional feature vectors that reflect the characteristics of the target user and the platform's activity status.

[0097] After obtaining the multi-dimensional feature vectors, this application can calculate the user value score of the target user based on these feature vectors. The user value score is a comprehensive indicator that considers not only multiple aspects of the target user, such as purchase history, browsing behavior, user preferences, and participation in platform activities, but also the industry characteristics of the e-commerce platform (such as promotional periods / flash sale scenarios) to assess the potential value and purchase intention of the target user. This application achieves "scenario-driven interface reconstruction" through deep coupling of user profiles and the industry characteristics of e-commerce platforms, overcoming industry problems such as static resource placements, cross-platform fragmentation, and ineffective commission incentives, and constructing a next-generation CPS display system with high conversion rates, low latency, and consistency across all platforms.

[0098] Furthermore, this application can employ various algorithms and models, such as logistic regression, decision trees, random forests, and neural networks, when calculating the user value score. These algorithms and models can be trained and learned based on multi-dimensional feature vectors to find features and patterns related to the target user value. By continuously adjusting and optimizing the parameters and structure of the algorithms and models, the accuracy and reliability of the user value score can be improved.

[0099] S130: Match target display strategies based on multi-dimensional data and user value score, and send target resource location data to terminal devices after obtaining the target resource location data according to the target display strategy.

[0100] In this step, after calculating the user value score rate of the target user based on the multi-dimensional feature vector in S120, this application can also match the target display strategy based on the multi-dimensional data and the user value score rate. In this way, the target resource position data can be obtained according to the target display strategy, and then the target resource position data can be sent to the terminal device so that the terminal device can dynamically display the front-end page according to the target resource position data.

[0101] Specifically, when matching target display strategies, this application comprehensively considers various factors from multi-dimensional data, such as the target user's purchase history, browsing behavior, user preferences, and the potential value and purchase intention reflected by the user value score. Simultaneously, this application also considers the current platform activity status, such as the type, scale, and duration of the activity, to ensure that the matched display strategy not only meets the user's personalized needs but also maximizes the use of platform resources to improve display effectiveness.

[0102] After matching is complete, this application will retrieve the corresponding target resource slot data from the resource slot database according to the target display strategy. This data includes, but is not limited to, product information, activity information, and promotional information, depending on the requirements of the target display strategy. For example, when the target display strategy includes a promotional resource slot, the business backend system can pull best-selling product data from CPS; when the target display strategy includes a team leader resource slot, the business backend system can query the team revenue API; when the target display strategy includes a new user resource slot, the business backend system can obtain exclusive products for new users; when the target display strategy includes a food expert resource slot, the business backend system can call LBS merchant services; when the target display strategy includes a student resource slot, the business backend system can obtain 3C digital products; when the target display strategy includes a high-value user resource slot, the business backend system can pull 3D product models.

[0103] After the retrieval is completed, this application can package the target resource location data into a specific format and send it to the terminal device via a network transmission protocol. Upon receiving this data, the terminal device will immediately initiate a dynamic rendering process to present the target resource location data to the user in an intuitive and attractive way, thereby guiding the user to make a purchase or participate in activities. For example, for commission-oriented users, the resource location will prioritize displaying sections such as high-commission best-selling products, activity order boards, sales ranking competitions, and earnings dashboards to meet their core need to earn commissions. For new user groups, the focus will be on displaying highly discounted sections such as free purchases and Taobao coupon purchases for 1 cent, to improve their first-time user experience and retention conversion rates. For users with specific interests, such as foodies, students, and stay-at-home mothers, the resource location will also be personalized according to their interests and preferences, displaying content that better meets their needs.

[0104] Furthermore, target applications can dynamically adjust based on the current activity status of the platform. For example, during promotional events, ad placements will prioritize displaying event-related sections to attract user attention and boost sales. Outside of promotional periods, the layout will be adjusted according to user behavior data and preferences.

[0105] This method of dynamically reconfiguring resource placements not only improves user experience and shopping efficiency but also helps e-commerce platforms achieve precise marketing and personalized services. By gaining a deeper understanding of the needs and preferences of target users, e-commerce platforms can more accurately push products and promotional information that match their needs, thereby increasing conversion rates and user stickiness. At the same time, this personalized service experience also helps enhance user loyalty to the e-commerce platform, thus significantly improving the platform's conversion efficiency.

[0106] In one embodiment, S110 collects user data of the target user bound to the terminal device and the current activity status of the platform, forming multi-dimensional data, which may include:

[0107] S111: Determine the target user bound to the terminal device.

[0108] S112: Quantify the team size of the target user's team through social relationship analysis thread.

[0109] S113: Obtain the user profile of the target user through the user profile recognition thread.

[0110] S114: Evaluate the user value level of the target user through the value evaluation thread.

[0111] S115: Determine whether the target user is a new user through the new customer determination thread.

[0112] S116: Detect the current platform's activity status through the promotion status detection thread.

[0113] S117: The current platform activity status, the target user's team size, user profile, user value level, and whether the target user is a new user are used as multi-dimensional data.

[0114] In this embodiment, the business backend system operates synchronously through various threads to improve the timeliness of data collection and processing. Specifically, the business backend system first identifies the target users bound to the terminal devices, ensuring the accuracy of subsequent analysis. The social relationship analysis thread provides data support for formulating team influence and incentive strategies by quantitatively analyzing the size of the target user's team. The user profile identification thread delves into the target user's personality traits, consumption behavior, and interests to build a detailed user profile. The value assessment thread accurately evaluates the target user's user value level based on the user profile and historical transaction records, laying the foundation for personalized recommendations and precision marketing. The new customer determination thread identifies whether the target user is a new user, providing a basis for formulating exclusive policies for new users and improving retention rates. The promotional status detection thread monitors the current platform's activity status in real time, ensuring that display strategies are closely aligned with platform activities.

[0115] The target application in this application has pre-established a comprehensive e-commerce promotion timeline synchronization system to ensure that the platform can accurately capture the marketing rhythm of major e-commerce platforms and provide users with timely and accurate promotional information. The specific implementation process is as follows:

[0116] I. E-commerce promotion timeline synchronization mechanism

[0117] Big Sales Promotion Data Collection and Preprocessing:

[0118] 1. Establish official data connection channels with major e-commerce platforms to obtain complete promotional schedules N days in advance;

[0119] 2. The operations team structured the data from the major promotion and clearly labeled it:

[0120] (1) Core promotional periods (such as pre-sale period, category day, peak period);

[0121] (2) Key promotional activities (such as cross-store discounts and limited-time flash sales);

[0122] (3) Special gameplay (such as interactive city, live broadcast exclusive discounts);

[0123] (4) Adjustment of commission policy (such as increased commission for specific product categories).

[0124] Intelligent data entry system:

[0125] 1. The R&D team has developed a dedicated backend for managing large-scale promotional activities, supporting:

[0126] (1) Visual timeline editing, supporting drag-and-drop activity scheduling;

[0127] (2) Automated interface mapping, associating major promotional activities with corresponding product APIs.

[0128] (3) Select the corresponding resource slot layout and binding according to the activity level and type. For example, special activities such as bonus red envelopes, super red carnival, and flash sale will be promoted on multiple resource slots such as the start page, homepage pop-up, full-screen flash sale and APP homepage. If it is a general activity, such as 200 minus 30, only the homepage pop-up and carousel resource slots can be generated for promotion.

[0129] 2. The system automatically generates an SQL script to write the configuration data into the special database table for the promotion and simultaneously refreshes it into the Redis cache.

[0130] II. Sources of User Profile Data

[0131] 1) Order Behavior Analysis: Through years of accumulated user order data, big data analysis and statistics of CPS channel order data are used to extract features such as product preferences, order frequency, and total commission amount;

[0132] 2) Team Expansion Tracking: Based on the invitation relationship chain, the number of direct / indirect downlines is counted to quantify team size;

[0133] 3) Device feature extraction: Capture the User-Agent field from the HTTP request header and parse the device model and OS version;

[0134] 4) User profile modeling: The XGBoost algorithm is used to build a user value model, which outputs consumption capacity tags and product promotion tendency predictions;

[0135] 5) Access the user feature data center to obtain pre-calculated tags: vertical interest tags (foodies / students, etc.) and price sensitivity scores.

[0136] III. User Value Assessment

[0137] The transaction center service uses a big data BI system to analyze the total amount and commission of a user's orders over the past N days, as well as the goods purchased, to assess the user's value. Users whose transaction volume and commission reach a certain level are classified as low / medium / high-value users.

[0138] IV. New User Determination

[0139] Verify the first order record in the order system and check the authorization status of third-party channels. If the user has not authorized or placed an order, they are considered a new user; otherwise, they are considered an existing user.

[0140] In the above embodiments, the data acquired and processed by these threads together constitute the core of multi-dimensional data, providing comprehensive and accurate data support for subsequent user value score calculation and display strategy matching.

[0141] In one embodiment, calculating the user value score rate of the target user based on the multidimensional feature vector in step S120 may include:

[0142] S121: The multidimensional feature vector is weighted using a time decay factor to obtain a multidimensional weighted feature vector.

[0143] S122: Input the multidimensional weighted feature vector into a pre-configured neural network model to obtain the user value score rate output by the neural network model.

[0144] In this embodiment, to more accurately assess the value of the target user, this application introduces a time decay factor. The time decay factor takes into account changes in user behavior and characteristics over time, assigning higher weights to newer data and lower weights to older data. In this way, the calculation of the user value score can more closely reflect the user's current state and needs.

[0145] After obtaining the multidimensional weighted feature vector, this application inputs it into a pre-configured neural network model for processing. This neural network model has undergone extensive data training and optimization, enabling it to accurately capture the relationship between the multidimensional feature vector and user value. After processing the input multidimensional weighted feature vector, the model outputs a user value score, which reflects the potential value and purchase intention of the target user.

[0146] By utilizing this method of calculating user value scores based on multi-dimensional feature vectors, this application enables accurate assessment of the value of target users. This not only helps e-commerce platforms formulate personalized display strategies, improve user experience and shopping efficiency, but also provides strong data support for e-commerce platforms' precision marketing and personalized services. Furthermore, by continuously optimizing the parameters and structure of the neural network model, this application can further improve the accuracy and reliability of the user value scores, laying a solid foundation for the long-term development of e-commerce platforms.

[0147] In one embodiment, weighting the multidimensional feature vector using a time decay factor in step S121 to obtain a multidimensional weighted feature vector may include:

[0148] S1211: Divide the multidimensional feature vector into a dynamic behavior region and a static attribute region.

[0149] S1212: The dynamic behavior region is weighted using exponential decay to obtain a weighted feature vector.

[0150] S1213: After combining the weighted feature vector with the static attribute region, a multidimensional weighted feature vector is obtained.

[0151] In this embodiment, to process multidimensional feature vectors more precisely, this application divides them into a dynamic behavior area and a static attribute area. The dynamic behavior area mainly includes data that changes over time, such as the user's purchase history and browsing behavior. This data can reflect the user's recent behavior and preferences. The static attribute area includes relatively stable information such as the user's age, gender, and region. This data is the user's basic attributes and is equally important for understanding user characteristics.

[0152] Furthermore, this application employs exponential decay for weighting the dynamic behavior region, which takes into account the changing trends of user behavior over time. The exponential decay function assigns different weights to features based on their proximity to the current moment. Behaviors more recent to the current moment receive greater weight and have a greater impact on the user value score. For example, the decay coefficient formula in this application (τ = 7 days as the half-life)...

[0153] The formula can be: `decay_factor = exp(-(current_timestamp - event_timestamp) / (7 * 86400))`, then the dynamic feature value update is: `weighted_value = raw_value * (1 + 3 * decay_factor)`, which means that the feature value weight has increased by 3 times in the past 7 days. In this way, this application can more accurately capture the user's recent behavior and preference changes, making the calculation of the user value score rate closer to the user's current state.

[0154] After obtaining the weighted feature vector, this application combines it with the static attribute region to form a multidimensional weighted feature vector. This multidimensional weighted feature vector contains both the user's dynamic behavioral information and the user's static attribute information, and can more comprehensively reflect the user's characteristics and needs. Inputting it into a neural network model for processing can yield a more accurate and reliable user value score, providing strong data support for personalized display and precision marketing on e-commerce platforms.

[0155] In one embodiment, such as Figure 2 As shown, Figure 2 This is a schematic diagram of the matching target display strategy provided in the embodiments of this application; S130, the matching target display strategy based on the multi-dimensional data and the user value score rate, may include:

[0156] S131: Determine the user type of the target user and the initial display strategy based on the score rate range corresponding to the user value score rate.

[0157] S132: Match multi-dimensional data with preset scenario rules to obtain matching results.

[0158] S133: Determine the target display strategy based on user type, initial display strategy, and matching results.

[0159] In this embodiment, during the matching of target display strategies, this application first categorizes target users into different user types based on their user value score and determines preliminary display strategies for them. This categorization helps e-commerce platforms develop differentiated display solutions for users of different values ​​to better meet their needs and preferences. For example, for high-value users, e-commerce platforms may display more high-end products and exclusive offers to enhance their shopping experience and loyalty. For low-value users, however, they may display more cost-effective products and promotional activities to attract their purchasing interest.

[0160] Next, this application can match multi-dimensional data with preset scenario rules. These scenario rules are formulated based on the operational experience of e-commerce platforms and user behavior data, aiming to recommend products and promotional information that best suit users' needs according to different user characteristics and scenario requirements. For example, when users are in the midst of a major promotional event, the scenario rules may prioritize recommending products and discounts related to the event to attract users' attention and boost sales. When users are in their daily shopping scenarios, personalized recommendations may be made based on their consumption habits and preferences.

[0161] Finally, this application determines the target display strategy based on user type, initial display strategy, and matching results. This process comprehensively considers multiple factors such as user value level, behavioral characteristics, and scenario needs, aiming to provide users with the display solution that best meets their needs. Through this precise matching method, e-commerce platforms can not only improve users' shopping experience and satisfaction, but also increase conversion rates and user stickiness, laying a solid foundation for the platform's long-term development.

[0162] In one embodiment, determining the user type of the target user and the initial display strategy based on the score rate range corresponding to the user value score rate in S131 may include:

[0163] S1311: If the user value score rate is within the first score rate range, then the user type of the target user is determined to be a high-value user, and the initial display strategy for the target user is the first combination strategy.

[0164] S1312: If the user value score rate is within the second score rate range, then the user type of the target user is determined to be a medium-value user, and the initial display strategy for the target user is the second combination strategy.

[0165] S1313: If the user value score rate is in the third score rate range, then the user type of the target user is determined to be a low-value user, and the initial display strategy for the target user is the default strategy.

[0166] In this embodiment, to more precisely meet the needs of users with different value levels, this application divides the user value score rate into three score rate intervals and determines the corresponding user type and initial display strategy for each interval. Specifically, when the user value score rate is in the first score rate interval, it indicates that the user has high purchasing potential and loyalty, and is therefore identified as a high-value user. For this type of user, the e-commerce platform can adopt a first combination strategy for display, which mainly focuses on displaying high-end products, exclusive offers, and personalized recommendations to improve the user's shopping experience and loyalty.

[0167] When a user's value score falls within the second score range, it indicates that the user has some purchasing potential and value, but it is slightly lower than that of high-value users, and therefore they are identified as medium-value users. For these users, e-commerce platforms can use a second combination strategy for display. This strategy balances product quality and price while also providing a degree of personalized recommendations based on the user's consumption habits and preferences to attract their purchasing interest and improve conversion rates.

[0168] When a user's value score falls within the third score range, it indicates that the user's purchasing potential and value are relatively low, thus classifying them as a low-value user. For these users, e-commerce platforms can employ a default display strategy, which primarily focuses on showcasing high-value products and promotional activities to attract user attention and boost sales.

[0169] By employing this differentiated display strategy, e-commerce platforms can better meet the needs of users with different values, enhance their shopping experience and satisfaction, and thus improve conversion rates and user stickiness.

[0170] In one embodiment, the multi-dimensional data may include the current activity status of the platform, the team size of the target user, the user profile, and whether the target user is a new user.

[0171] In step S132, the multi-dimensional data is matched with preset scene rules to obtain a matching result, which may include:

[0172] S1321: When the current platform's activity status is a major promotion, and the target user's team size exceeds a preset size threshold, the intermediate display strategy for the target user is determined to be the first combination strategy.

[0173] S1322: Otherwise, when the user profile of the target user matches a specific tag, the intermediate display strategy for the target user is determined to be the second combination strategy.

[0174] S1323: When the user profile of the target user does not match the specific tag, but the target user is a new user, the intermediate display strategy for the target user is determined to be the third combination strategy.

[0175] S1324: Otherwise, determine the intermediate display strategy for the target user as the default strategy.

[0176] In this embodiment, during the matching of target display strategies, this application further considers multi-dimensional data such as the current activity status of the platform, the team size of the target user, user profile, and whether the target user is a new user. The comprehensive use of this data makes the matching of display strategies more accurate and personalized.

[0177] Specifically, when the platform is currently undergoing a major promotional event, it signifies that the platform is carrying out significant promotional activities, making it crucial to attract user participation and purchases. If the target user's team size exceeds a preset threshold, it indicates that the user possesses considerable social influence and team resources. Therefore, this application determines the target user's intermediate display strategy as the first combination strategy. This strategy focuses on showcasing promotional activities related to team expansion, such as rewards for inviting friends, to incentivize users to utilize their team resources to participate in the activities, thereby expanding the event's influence and engagement.

[0178] If the platform is not currently engaged in a major promotional event, or if the target user's team size does not reach a preset threshold, this application will further consider the target user's profile. When the user profile matches specific tags, such as food enthusiasts or students, it indicates that the user has specific interests and consumption tendencies. Therefore, this application can determine that the intermediate display strategy for the target user is the second combination strategy. This strategy may focus on displaying products and promotional information related to the user profile tags, such as food festival promotions for food enthusiasts and discounts on educational products for students, to meet the user's specific needs and interests.

[0179] If the target user's profile does not match a specific tag, but the target user is a new user, this application can determine that the intermediate display strategy for the target user is a third combination strategy. This strategy may focus on displaying exclusive offers and guidance information for new users, such as a red envelope for new user registration or a discount on the first order for new users, in order to attract new users to participate in purchases and increase their stickiness and loyalty to the platform.

[0180] If the target user does not match specific user profile tags and is not a new user, this application can determine that the intermediate display strategy for the target user is the default strategy. This strategy may focus on displaying the platform's regular promotional activities and popular products to attract the user's attention and promote sales.

[0181] By comprehensively considering multi-dimensional data matching, this application can provide users with more personalized display strategies that meet their needs, thereby improving users' shopping experience and satisfaction, and laying a solid foundation for the long-term development of e-commerce platforms.

[0182] In one embodiment, the user type may include high-value users, medium-value users, and low-value users.

[0183] The initial demonstration strategy may include a first combination strategy, a second combination strategy, and a third combination strategy.

[0184] The matching results may include a first combination strategy, a second combination strategy, a third combination strategy, and a default strategy.

[0185] In step S133, determining the target display strategy based on the user type, the preliminary display strategy, and the matching result may include:

[0186] S1331: When the user type is a high-value user, the initial display strategy is the first combination strategy, but the matching result is the default strategy, the second combination strategy is used as the target display strategy.

[0187] S1332: When the user type is a medium-value user, the initial display strategy is the second combination strategy, but the matching result is the default strategy, the third combination strategy is used as the target display strategy.

[0188] S1333: Otherwise, the matching result shall be used as the target display strategy.

[0189] In this embodiment, the application also considers handling some special cases during the determination of the target display strategy. Specifically, when the user type is a high-value user and the initial display strategy is the first combination strategy, this usually means that the e-commerce platform wants to provide the highest-end display solution for high-value users. However, if, based on the matching results of multi-dimensional data, the user is not suitable for the first combination strategy and instead receives the default strategy, this application will not directly adopt the default strategy, but will instead use the second combination strategy as the target display strategy. This failure degradation mechanism aims to ensure that even among high-value users, adjustments can be made flexibly according to their specific circumstances, avoiding overly rigid or unsuitable display solutions.

[0190] Similarly, when the user type is a mid-value user and the initial display strategy is the second combination strategy, if the matching result is the default strategy, this application will use the third combination strategy as the target display strategy. This approach aims to provide mid-value users with a display solution that better matches their purchasing potential and value, avoiding overly high-end or low-end display strategies that may cause them discomfort or loss of interest.

[0191] For other cases, this application directly uses the matching results as the target display strategy. This ensures both the personalization and accuracy of the display strategy while avoiding overly complex and unnecessary processing procedures. Furthermore, this application can dynamically allocate weights so that users with high scores can automatically trigger high-value resource positions (such as a 50% increase in the exposure weight of the earnings dashboard).

[0192] By comprehensively considering user type, initial display strategy, and matching results, this application can provide users with a display solution that better matches their needs and preferences. This not only enhances the user's shopping experience and satisfaction but also increases the conversion rate and user stickiness of the e-commerce platform, laying a solid foundation for the platform's long-term development.

[0193] In one embodiment, such as Figure 3 As shown, Figure 3 This application provides a flowchart illustrating a method for dynamically displaying resource bits according to an embodiment of the present application; the present application also provides a method for dynamically displaying resource bits, which is applied to a terminal device and may include:

[0194] S210: Send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system.

[0195] S220: Dynamically render the front-end page based on the target resource location data, and display the rendered front-end page to the target user.

[0196] In this embodiment, the terminal device can obtain resource slot data related to the target user in real time by sending a resource slot data acquisition request to the business backend system. This data may include product recommendation information, details of promotional activities, and product categories that the user is interested in. After receiving the request, the business backend system will generate resource slot data suitable for the user based on preset rules and algorithms, combined with the target user's user profile, historical behavior data, and the current platform's operation strategy, and return it to the terminal device.

[0197] After receiving the target resource location data returned by the business backend system, this application will dynamically render the front-end page based on this data. This process may involve multiple aspects such as adjusting the page layout, displaying product information, and highlighting promotional activities. Through dynamic rendering, it can be ensured that the content displayed on the front-end page is highly consistent with the needs and preferences of the target users, thereby improving the user's shopping experience and satisfaction.

[0198] Finally, this application displays the rendered front-end page to the target user. At this point, the user can see customized product recommendations, promotional activities, and other information. This information not only aligns with the user's interests and needs but also stimulates their desire to purchase, promoting sales conversion. In this way, this application achieves dynamic display of resource slots, providing strong support for personalized marketing on e-commerce platforms.

[0199] In one embodiment, such as Figure 4 As shown, Figure 4 This is a schematic diagram illustrating the process of dynamically rendering a front-end page according to an embodiment of this application; S220, which involves dynamically rendering the front-end page based on the target resource bit data, may include:

[0200] S221: Invoke the main thread to initialize the infrastructure based on the target resource bit data;

[0201] S222: After the main thread is initialized, the user-specific layout calculation task, network image prefetching task, local cache reading task and device performance detection task are executed in parallel according to the target resource bit data through the parallel asynchronous loading pool, and the page skeleton is constructed through the main thread.

[0202] S223: Perform preemptive rendering on the page skeleton based on the task status of parallel tasks until all resource slots are loaded.

[0203] In this embodiment, efficient multi-threading technology is employed during the dynamic rendering of the front-end page. Specifically, the main thread is first invoked to initialize the infrastructure based on the target resource location data. This is the first step in building the page, providing the necessary framework and support for subsequent operations.

[0204] After the main thread initializes, to improve rendering efficiency and user experience, this application uses a parallel asynchronous loading pool to execute multiple tasks in parallel. These tasks include user-specific layout calculations, network image prefetching, local cache retrieval, and device performance testing. The parallel execution of these tasks fully utilizes the device's processing power, shortening page loading and rendering time.

[0205] Meanwhile, the main thread is also building the page skeleton. The page skeleton is the basic structure of the page, providing users with an initial view of the page so they can begin browsing before all resource data is fully loaded.

[0206] During parallel task execution, this application can also perform preemptive rendering on the page skeleton based on the task's status. This means that once a task completes the loading and processing of the resource data it is responsible for, it will immediately render it into the page skeleton. This preemptive rendering method ensures that the page can display complete content as quickly as possible, improving the user's waiting experience. In addition, since each resource in this application has a corresponding skeleton screen of a certain shape, the front-end page displays the skeleton screen before the resource is fully rendered. Once resources such as images are downloaded, the loaded resources are displayed.

[0207] Furthermore, when multiple resource slots are loaded simultaneously, this application can render and display them according to priority. For example, the splash screen resource slot and the homepage pop-up have the highest priority and are rendered according to their priority. This process continues until all resource slots are loaded. At this point, the user will see a fully rendered front-end page that meets their needs and preferences. This page not only displays products and promotional activities that the user is interested in, but also enhances the user's shopping experience and satisfaction through careful layout and design.

[0208] Furthermore, such as Figure 5 As shown, Figure 5 This is a schematic diagram illustrating the three-tier unified architecture implemented using the Flutter framework in this application. This application can implement a three-tier unified architecture based on the Flutter framework, abandoning the old multi-platform development model, unifying the Flutter development interface, and enabling multiple platforms to share a single codebase. Specifically, when using the Flutter framework, this application can first initialize the Dart VM, Skia graphics engine, etc., and then build the basic structure and layout of the page. This three-tier unified architecture can include a presentation layer, a business logic layer, and a data access layer. The presentation layer is mainly responsible for page display and user interaction; it can dynamically update the page content and layout based on user behavior and feedback. The business logic layer encapsulates the core business logic of the e-commerce platform, such as user profile construction and display strategy matching. It can execute corresponding business operations based on user information and requests from the presentation layer and return the results to the presentation layer for display. The data access layer is responsible for interaction with the database or backend system; it can query or update data from the database according to the needs of the business logic layer, ensuring data accuracy and consistency.

[0209] This three-tiered unified architecture allows multiple platforms to share a single codebase, significantly improving development efficiency and code maintainability. Furthermore, the separation of the presentation layer, business logic layer, and data access layer enables independent development and testing of each part of the system, reducing system coupling and complexity. In addition, the Flutter framework-based implementation ensures consistency and smoothness across different platforms, enhancing the user experience.

[0210] During implementation, developers can use the Dart language for coding and leverage the rich components and APIs provided by the Flutter framework to quickly build e-commerce applications with abundant functionality and a good user experience. Simultaneously, by designing the business logic layer and data access layer appropriately, efficient application operation and accurate data processing can be ensured. During the testing phase, the hot reload and hot restart features provided by the Flutter framework can be used to quickly iterate and fix errors and problems in the code, thereby effectively improving development efficiency and quality.

[0211] In this way, this application achieves dynamic, efficient, and personalized display of resource slots, providing strong support for the operation and development of e-commerce platforms.

[0212] In one embodiment, S222 may include the parallel asynchronous loading pool executing user-specific layout calculation tasks, network image prefetching tasks, local cache reading tasks, and device performance detection tasks in parallel based on the target resource bit data, and may include:

[0213] S2221: Execute user-specific layout calculation tasks and device performance detection tasks based on the target resource bit data through a parallel asynchronous loading pool.

[0214] S2222: The detection results of the device performance detection task are fed back to the decision engine that executes the user-specific layout calculation task through the parallel asynchronous loading pool.

[0215] S2223: After the decision engine returns the layout data, a local cache read task is triggered through the parallel asynchronous loading pool, and the resource bit version in the local cache is compared with the resource bit version in the target resource bit data to obtain the comparison result.

[0216] S2224: After determining the resources to be downloaded based on the comparison results through the parallel asynchronous loading pool, the network image prefetching task is triggered to download the resources.

[0217] In this embodiment, during the execution of tasks in the parallel asynchronous loading pool, this application pays special attention to the user-specific layout calculation task and the device performance testing task. The user-specific layout calculation task aims to calculate the most suitable page layout for the user based on the target resource bit data and the user's preferences. This task needs to fully consider factors such as the user's screen size, resolution, and usage habits to ensure that the page layout is both aesthetically pleasing and practical. The device performance testing task, on the other hand, comprehensively evaluates the user's device performance, including key indicators such as CPU, memory, network speed, and device model. The purpose of this task is to provide a strong basis for subsequent resource loading and rendering, ensuring that the page can run smoothly on the user's device.

[0218] Understandably, the purpose of this application in detecting device performance is that some resource slots are performance-intensive, such as short videos. If the current terminal device is a low-end device, it may cause page lag and affect user experience. Therefore, when this application detects such devices, it can feed this information into the user-specific layout calculation task so that the original short video can be replaced with a static image display.

[0219] Furthermore, to effectively apply the device performance testing results to user-specific layout calculations, this application can use a parallel asynchronous loading pool to feed the testing results back to the decision engine executing the user-specific layout calculation task in real time. Upon receiving this feedback, the decision engine will adjust and optimize the user-specific layout calculation task according to the device's performance. For example, in cases of poor device performance, the decision engine may choose a simpler and more efficient layout scheme to reduce the consumption of device resources and improve page loading speed and responsiveness.

[0220] Once the user-specific layout calculation task is complete, the decision engine returns the layout data to the parallel asynchronous loading pool. At this point, the parallel asynchronous loading pool triggers a local cache read task, comparing the resource bit version in the locally cached resource bit data with the resource bit version in the target resource bit data. The purpose of this comparison is to determine whether a new resource needs to be downloaded from the network. If the locally cached resource bit version matches the version in the target resource bit data, the locally cached resource can be used directly, reducing network requests and loading time. If the versions do not match, or the version is not available, a network image prefetch task needs to be triggered to download the new resource from the network.

[0221] During the execution of the network image prefetching task, this application also fully considers factors such as network conditions and user needs. For example, when the network speed is slow, this application may prioritize downloading the resources that the user is most interested in or that are most important to the user, to ensure that the user can see the content they want to see as quickly as possible. At the same time, this application will also intelligently cache and manage the downloaded resources so that these resources can be quickly obtained in subsequent use, thereby improving the user experience.

[0222] The following describes the resource bit dynamic reconstruction and display device provided in the embodiments of this application. The resource bit dynamic reconstruction and display device described below can be referred to in correspondence with the resource bit dynamic reconstruction and display method described above.

[0223] In one embodiment, such as Figure 6 As shown, Figure 6 This is a schematic diagram of a resource bit dynamic reconstruction device provided in an embodiment of this application. This application also provides a resource bit dynamic reconstruction device, which may include a data acquisition module 110, a value calculation module 120, and a dynamic reconstruction module 130, specifically including the following:

[0224] The data acquisition module 110 is used to respond to the resource bit data acquisition request sent by the terminal device, collect the user data of the target user bound to the terminal device and the current activity status of the platform, and form multi-dimensional data.

[0225] The value calculation module 120 is used to convert the multi-dimensional data into a multi-dimensional feature vector and then calculate the user value score rate of the target user based on the multi-dimensional feature vector.

[0226] The dynamic reconstruction module 130 is used to match the target display strategy based on the multi-dimensional data and the user value score rate, and to obtain the target resource position data according to the target display strategy and send it to the terminal device.

[0227] In the above embodiments, when the business backend system receives a resource slot data acquisition request sent by the terminal device, it can collect diverse user data of the target user bound to the terminal device, including but not limited to user behavior data and user preference data, and combine this data with the current platform activity status to form multi-dimensional data. This multi-dimensional data is further transformed into multi-dimensional feature vectors, and the user value score rate of the target user is calculated based on these vectors. This user value score rate can intuitively reflect the user's potential value and purchase intention, providing a scientific basis for formulating differentiated display strategies. Furthermore, based on this multi-dimensional data and user value score rate, this application can intelligently match the most suitable target display strategy for the target user, and accordingly acquire the target resource slot data and send it to the terminal device for dynamic rendering and display. This series of dynamic reconstruction and personalized display processes not only greatly improves the user experience but also effectively increases user retention and conversion efficiency, bringing new development opportunities to e-commerce applications.

[0228] In one embodiment, such as Figure 7 As shown, Figure 7 This application provides a schematic diagram of a resource bit dynamic display device according to an embodiment of the present application; the present application also provides a resource bit dynamic display device, which may include a data receiving module 210 and a dynamic display module 220, specifically including the following:

[0229] The data receiving module 210 is used to send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system.

[0230] The dynamic display module 220 is used to dynamically render the front-end page according to the target resource location data and display the rendered front-end page to the target user.

[0231] In the above embodiments, after receiving the target resource location data returned by the business backend system, the terminal device dynamically renders the front-end page based on this data. This process may involve multiple aspects such as adjusting the page layout, displaying product information, and highlighting promotional activities. Through dynamic rendering, it can be ensured that the content displayed on the front-end page is highly consistent with the needs and preferences of the target users, thereby improving the user's shopping experience and satisfaction.

[0232] Next, this application can display the rendered front-end page to the target user. At this point, the user can see customized product recommendations, promotional activities, and other information. This information not only matches the user's interests and needs but also stimulates their desire to purchase, promoting sales conversion. In this way, this application achieves dynamic display of resource slots, providing strong support for personalized marketing on e-commerce platforms.

[0233] In one embodiment, this application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the resource bit dynamic reconstruction method as described in any of the above embodiments, and / or the steps of the resource bit dynamic display method as described in any of the above embodiments.

[0234] In one embodiment, this application also provides a computer device, including: one or more processors, and memory.

[0235] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the resource bit dynamic reconstruction method as described in any of the above embodiments, and / or the steps of the resource bit dynamic display method as described in any of the above embodiments.

[0236] Indicatively, such as Figure 8 As shown, Figure 8 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 8 The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions executable by the processing component 302, such as application programs. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform the resource bit dynamic reconstruction and display method of any of the above embodiments.

[0237] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0238] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0239] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, 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, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0240] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0241] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for dynamic reconfiguration of resource bits, characterized in that, The method is applied to a business back-end system, including: In response to a resource bit data acquisition request sent by a terminal device, the system collects user data of the target user bound to the terminal device and the current activity status of the platform, and forms multi-dimensional data. After converting the multi-dimensional data into a multi-dimensional feature vector, the user value score rate of the target user is calculated based on the multi-dimensional feature vector. Based on the multi-dimensional data and the user value score, a target display strategy is matched, and target resource location data is obtained according to the target display strategy and then sent to the terminal device.

2. The resource bit dynamic reconstruction method according to claim 1, characterized in that, The process involves collecting user data of the target user bound to the terminal device and the current activity status of the platform, forming multi-dimensional data, including: Identify the target user bound to the terminal device; The team size of the target user's team is quantified and statistically analyzed through social relationship analysis. The user profile of the target user is obtained through the user profile recognition thread; The target user's user value level is assessed through a value assessment thread; The new customer determination thread determines whether the target user is a new user. The current activity status of the platform is detected through a special promotion status detection thread; The current platform activity status, the target user's team size, user profile, user value level, and whether the target user is a new user are considered as multi-dimensional data.

3. The resource bit dynamic reconstruction method according to claim 1, characterized in that, The step of calculating the user value score rate of the target user based on the multidimensional feature vector includes: The multidimensional feature vector is weighted using a time decay factor to obtain a multidimensional weighted feature vector; The multidimensional weighted feature vector is input into a pre-configured neural network model to obtain the user value score rate output by the neural network model.

4. The resource bit dynamic reconstruction method according to claim 1, characterized in that, The step of weighting the multidimensional feature vector using a time decay factor to obtain a multidimensional weighted feature vector includes: The multidimensional feature vector is divided into a dynamic behavior region and a static attribute region; The dynamic behavior region is weighted using exponential decay to obtain a weighted feature vector; The weighted feature vector is combined with the static attribute region to obtain a multidimensional weighted feature vector.

5. The resource bit dynamic reconstruction method according to claim 1, characterized in that, The target display strategy based on the multi-dimensional data and the user value score includes: The user type of the target user and the initial display strategy are determined based on the score rate range corresponding to the user value score rate. The multi-dimensional data is matched with preset scene rules to obtain matching results; The target display strategy is determined based on the user type, the preliminary display strategy, and the matching results.

6. The resource bit dynamic reconstruction method according to claim 5, characterized in that, The step of determining the user type of the target user and the initial display strategy based on the score rate range corresponding to the user value score rate includes: If the user value score rate is within the first score rate range, then the user type of the target user is determined to be a high-value user, and the initial display strategy for the target user is the first combination strategy. If the user value score rate is within the second score rate range, then the user type of the target user is determined to be a medium-value user, and the initial display strategy for the target user is the second combination strategy. If the user value score rate is in the third score rate range, then the user type of the target user is determined to be a low-value user, and the initial display strategy for the target user is the default strategy.

7. The resource bit dynamic reconstruction method according to claim 5, characterized in that, The multi-dimensional data includes the current activity status of the platform, the team size of the target user, the user profile, and whether the target user is a new user. The step of matching the multi-dimensional data with preset scene rules to obtain matching results includes: When the current platform activity status is a major promotion status, and the team size of the target user exceeds a preset size threshold, the intermediate display strategy for the target user is determined to be the first combination strategy; Otherwise, if the user profile of the target user matches a specific tag, the intermediate display strategy for the target user is determined to be the second combination strategy; When the user profile of the target user does not match the specific tag, but the target user is a new user, the intermediate display strategy for the target user is determined to be the third combination strategy; Otherwise, the intermediate display strategy for the target user is determined to be the default strategy.

8. The resource bit dynamic reconstruction method according to claim 5, characterized in that, The user types include high-value users, medium-value users, and low-value users; The initial demonstration strategy includes a first combination strategy, a second combination strategy, and a third combination strategy; The matching results include a first combination strategy, a second combination strategy, a third combination strategy, and a default strategy; The step of determining the target display strategy based on the user type, the preliminary display strategy, and the matching result includes: When the user type is a high-value user, the initial display strategy is the first combination strategy, but the matching result is the default strategy, the second combination strategy is used as the target display strategy. When the user type is a medium-value user, the initial display strategy is the second combination strategy, but the matching result is the default strategy, the third combination strategy is used as the target display strategy. Otherwise, the matching result will be used as the target display strategy.

9. A method for dynamically displaying resource slots, characterized in that, The method is applied to a terminal device and includes: Send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system; The front-end page is dynamically rendered based on the target resource location data, and the rendered front-end page is displayed to the target user.

10. The method for dynamically displaying resource slots according to claim 9, characterized in that, The step of dynamically rendering the front-end page based on the target resource location data includes: The main thread is invoked to initialize the infrastructure based on the target resource bit data. After the main thread is initialized, the user-specific layout calculation task, network image prefetching task, local cache reading task and device performance detection task are executed in parallel according to the target resource bit data through the parallel asynchronous loading pool, and the page skeleton is constructed through the main thread. Preemptive rendering is performed on the page skeleton based on the task status of parallel tasks until all resource slots are loaded.

11. The method for dynamically displaying resource slots according to claim 10, characterized in that, The process of executing user-specific layout calculation tasks, network image prefetching tasks, local cache reading tasks, and device performance detection tasks in parallel using a parallel asynchronous loading pool based on the target resource bit data includes: The user-specific layout calculation task and device performance detection task are executed by a parallel asynchronous loading pool based on the target resource bit data. The detection results of the device performance detection task are fed back to the decision engine that executes the user-specific layout calculation task through the parallel asynchronous loading pool; After the decision engine returns the layout data, it triggers a local cache read task through the parallel asynchronous loading pool, and compares the resource bit version in the local cache with the resource bit version in the target resource bit data to obtain the comparison result. After determining the resources to be downloaded based on the comparison results using the parallel asynchronous loading pool, a network image prefetching task is triggered to download the resources.

12. A resource bit dynamic reconfiguration device, characterized in that, include: The data acquisition module is used to respond to the resource bit data acquisition request sent by the terminal device, collect the user data of the target user bound to the terminal device and the current activity status of the platform, and form multi-dimensional data; The value calculation module is used to convert the multi-dimensional data into a multi-dimensional feature vector, and then calculate the user value score rate of the target user based on the multi-dimensional feature vector. The dynamic reconstruction module is used to match the target display strategy based on the multi-dimensional data and the user value score, and to obtain the target resource position data according to the target display strategy and send it to the terminal device.

13. A dynamic resource location display device, characterized in that, include: The data receiving module is used to send a resource bit data acquisition request to the business backend system and receive the target resource bit data returned by the business backend system. The dynamic display module is used to dynamically render the front-end page based on the target resource location data and display the rendered front-end page to the target user.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the resource bit dynamic reconstruction method as described in any one of claims 1 to 8, and / or the steps of the resource bit dynamic display method as described in any one of claims 9 to 11.

15. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the resource bit dynamic reconstruction method as described in any one of claims 1 to 8, and / or the steps of the resource bit dynamic display method as described in any one of claims 9 to 11.

Images