Display page generation method and device, computer device and storage medium

Abstract

The application relates to a display page generation method and device, computer equipment and a storage medium. The method comprises the following steps: in response to a page entry triggering operation, obtaining basic data and behavior data of a target user; performing feature extraction on the basic data and the behavior data to obtain target feature information; performing interaction willingness decision on the target feature information based on a decision tree model to obtain a user interaction type reflecting the interaction willingness of the target user to a product; the decision tree model is obtained by model construction based on user features in a user feature sample set reaching a modeling condition; based on the user interaction type, determining a task to be participated and a task display sequence of each task; a reward corresponding to the task is used for interacting with the target product; and according to the task, the task display sequence and the target product, a display page matched with the user interaction type is generated. The method can improve the product interaction rate.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer device, and storage medium for generating display pages. Background Technology

[0002] With the development of internet technology, more and more users are interacting with products through smart terminals, such as purchasing goods or handling related business. Therefore, when users visit a website or client and enter the display page, product providers usually display some recommended products and interactive promotional activities for those products. In traditional solutions, different types of users often display the same display page when visiting a website or client, which is not conducive to improving product interaction rates. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, apparatus, computer device, and storage medium for generating display pages that can help improve product interaction rates, in order to address the aforementioned technical problems.

[0004] A method for generating a display page, the method comprising:

[0005] In response to actions triggered by page entry points, obtain basic and behavioral data of the target user;

[0006] Feature extraction is performed on the basic data and the behavioral data to obtain target feature information;

[0007] The decision tree model is used to make decisions on the interaction willingness of the target feature information to obtain the user interaction type that reflects the target user's willingness to interact with the product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions.

[0008] Based on the user interaction type, the tasks to be participated in and the task display order of each task are determined; the rewards corresponding to the tasks are used to interact with the target product.

[0009] Based on the task, the task display order, and the target product, a display page matching the user interaction type is generated.

[0010] A display page generation apparatus, the apparatus comprising:

[0011] The acquisition module is used to acquire basic and behavioral data of the target user in response to page entry trigger operations;

[0012] The extraction module is used to extract features from the basic data and the behavioral data to obtain target feature information;

[0013] The decision module is used to make decisions on the interaction willingness based on the target feature information based on the decision tree model, and to obtain the user interaction type that reflects the target user's interaction willingness with the product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions;

[0014] The determining module is used to determine the tasks to be participated in and the task display order of each task based on the user interaction type; the reward corresponding to the task is used to interact with the target product;

[0015] The generation module is used to generate a display page that matches the user interaction type based on the task, the task display order, and the target product.

[0016] In one embodiment, the device further includes:

[0017] The calculation module is used to calculate the information entropy of each user feature in the user feature sample set to obtain a first information entropy; after dividing the user feature sample set into subsets according to each user feature in the user feature sample set, it calculates the second information entropy corresponding to each subset; and based on the first information entropy and the second information entropy, it determines the information gain of each user feature in the user feature sample set.

[0018] In one embodiment, the device further includes:

[0019] A construction module is used to acquire a user feature sample set; select user features whose information gain meets the modeling conditions from the user feature sample set to obtain a user feature sample subset composed of user features whose information gain meets the modeling conditions; take the user feature with the largest information gain from the user feature sample subset as the root node; use the root node as the judgment condition to judge user attributes or user behaviors to obtain the target branch; select the user feature with the second largest information gain from the user feature sample subset as the node of the target branch, until the leaf node is reached to complete the construction of the decision tree model.

[0020] In one embodiment, the construction module is further configured to sort the user features in the feature sample set in descending order according to the magnitude of the information gain; select the user features ranked first in the sorted feature sample set; and combine the selected user features into a subset of user feature samples.

[0021] In one embodiment, the user interaction type includes a first interaction type where the degree of willingness to interact with the product is greater than a first threshold and a second interaction type where the degree of willingness is less than a second threshold; the device further includes:

[0022] The display module is configured to generate a first display page matching the first user interaction type based on the task, the task display order, and the target product when the user interaction type is the first interaction type; and to generate a second display page matching the second user interaction type based on the task, the task display order, and the target product when the user interaction type is the second interaction type; wherein the task display order configured in the first display page is different from the task display order configured in the second display page; and / or, at least one of the tasks configured in the first display page and the tasks configured in the second display page is different.

[0023] In one embodiment, the display page is configured with at least two target tasks; the device further includes:

[0024] The processing module is used to respond to task processing operations by sequentially entering the task pages of each target task; when the corresponding target task is completed on the task page, the reward for the corresponding target task is obtained; and the product in the display page is interacted with based on the reward.

[0025] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0026] In response to actions triggered by page entry points, obtain basic and behavioral data of the target user;

[0027] Feature extraction is performed on the basic data and the behavioral data to obtain target feature information;

[0028] The decision tree model is used to make decisions on the interaction willingness of the target feature information to obtain the user interaction type that reflects the target user's willingness to interact with the product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions.

[0029] Based on the user interaction type, the tasks to be participated in and the task display order of each task are determined; the rewards corresponding to the tasks are used to interact with the target product.

[0030] Based on the task, the task display order, and the target product, a display page matching the user interaction type is generated.

[0031] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0032] In response to actions triggered by page entry points, obtain basic and behavioral data of the target user;

[0033] Feature extraction is performed on the basic data and the behavioral data to obtain target feature information;

[0034] The decision tree model is used to make decisions on the interaction willingness of the target feature information to obtain the user interaction type that reflects the target user's willingness to interact with the product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions.

[0035] Based on the user interaction type, the tasks to be participated in and the task display order of each task are determined; the rewards corresponding to the tasks are used to interact with the target product.

[0036] Based on the task, the task display order, and the target product, a display page matching the user interaction type is generated.

[0037] The aforementioned method, apparatus, computer equipment, and storage medium for generating display pages extract features from basic and behavioral data. Based on a decision tree model, they determine the user's willingness to interact using the extracted target feature information, resulting in a user interaction type reflecting the target user's willingness to interact with the product. Based on the user interaction type, they determine the tasks to be participated in and the task display order. According to the tasks, task display order, and target product, they generate display pages matching the user interaction type. This allows for different display pages to be generated for users with different interaction types, and targeted settings for different tasks and task display orders for different users. This enables users to participate in corresponding tasks and receive rewards to interact with the product, thus improving the product's interaction rate. Furthermore, the decision tree model is constructed based on user features whose information gain meets the modeling conditions within the user feature sample set. This makes the constructed decision tree model more accurate in judging user interaction types, resulting in more precise display pages displayed to users. This effectively encourages users to participate in corresponding tasks and receive rewards to interact with the product, further improving the product's interaction rate. Attached Figure Description

[0038] Figure 1 This is a diagram illustrating the application environment of a page generation method in one embodiment.

[0039] Figure 2 This is a flowchart illustrating a page generation method in one embodiment;

[0040] Figure 3 This is a schematic diagram of the structure of a decision tree model in one embodiment;

[0041] Figure 4 This is a schematic diagram illustrating the order in which different tasks are displayed in another embodiment;

[0042] Figure 5 This is a schematic diagram of the display page corresponding to the first user interaction type in one embodiment;

[0043] Figure 6 This is a schematic diagram of the display page corresponding to the second user interaction type in one embodiment;

[0044] Figure 7 This is a schematic diagram of the display page corresponding to the first user interaction type in another embodiment;

[0045] Figure 8 This is a schematic diagram of the display page corresponding to the second user interaction type in another embodiment;

[0046] Figure 9 This is a flowchart illustrating the steps involved in constructing a decision tree model in one embodiment.

[0047] Figure 10 This is a structural block diagram illustrating a page generation apparatus in one embodiment;

[0048] Figure 11 This is a structural block diagram of a page generation apparatus shown in another embodiment;

[0049] Figure 12 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0051] The display page generation method provided in this application can be applied to, for example... Figure 1The application environment shown includes a terminal 102 and a server 104. In response to a page entry trigger operation, the terminal 102 obtains basic and behavioral data of the target user from the server 104 or locally; it extracts features from the basic and behavioral data to obtain target feature information; based on a decision tree model, it makes a decision on the target feature information regarding the user's willingness to interact, resulting in a user interaction type reflecting the target user's willingness to interact with the product; the decision tree model is constructed based on user features whose information gain meets the modeling conditions within the user feature sample set; based on the user interaction type, it determines the tasks to be participated in and the task display order; the reward corresponding to the task is used to interact with the target product; and based on the task, the task display order, and the target product, it generates a display page matching the user interaction type. The user feature sample set can be obtained from the server 104, and the decision tree model can be constructed by the terminal 102 or by the server 104 and deployed on the terminal 102.

[0052] The terminal 102 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these.

[0053] Server 104 can be an independent physical server or a service node in a blockchain system. The service nodes in the blockchain system form a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol that runs on top of the Transmission Control Protocol (TCP).

[0054] In addition, server 104 can also be a server cluster consisting of multiple physical servers, which can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0055] Terminal 102 and server 104 can be connected via Bluetooth, USB (Universal Serial Bus) or network, etc., and this application does not impose any restrictions.

[0056] In one embodiment, such as Figure 2 As shown, a method for generating a display page is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0057] S202, responding to the page entry trigger operation, obtain the target user's basic data and behavioral data.

[0058] The page entry trigger operation can be triggered by the entry control in the product entry page, such as the entry control of insurance products, wealth management products and other products or services in the product entry page.

[0059] Basic data can be fundamental characteristic information of the target user, including their age, gender, residential address, hobbies, occupation, income, education level, smart devices used, and network address. Behavioral data can be characteristic information of the target user's behavior, including whether the target user has paid to become a member (i.e., whether they are a paying user) on the product's interactive website, client, and / or mini-program, their first visit time, number of visits, and number of tasks participated in. The number of visits could be the number of times the user visits the website, client, and / or mini-program in a month, or the number of times they visit the product in a week.

[0060] In one embodiment, the terminal displays a product entry page. When a page entry trigger operation is detected by an entry control on the product entry page, basic data and behavioral data are obtained based on the target user's user identifier. This user identifier can be the user's username or login account on the website, client, and / or mini-program where the product is interacted with.

[0061] S204. Feature extraction is performed on the basic data and behavioral data to obtain target feature information.

[0062] Among them, target feature information can be key feature information of the target user, including basic feature information and behavioral feature information of the target user.

[0063] In one embodiment, a terminal can use a feature extraction tool to extract target feature information from basic data and behavioral data, such as the target user's gender, age, number of visits, first visit time, and whether they are a paying user. This feature extraction tool can be configured with extraction parameters to determine which feature information needs to be extracted.

[0064] S206, Based on the decision tree model, make a decision on the interaction willingness of the target feature information to obtain the user interaction type that reflects the target user's willingness to interact with the product.

[0065] Interaction intention refers to the target user's willingness to interact with the product, such as whether the target user intends to purchase a certain product, or the degree of their intention to purchase the product. This decision tree model is constructed based on user features within the user feature sample set whose information gain meets the modeling conditions.

[0066] In one embodiment, the terminal inputs target feature information into a decision tree model, and makes user attribute or user behavior judgments on the target feature information based on the root node of the decision tree model, thereby determining the branch to which the target feature information flows. Then, based on the node of the branch, it makes user attribute or user behavior judgments on the target feature information to determine the next branch to which the target feature information flows, until it reaches the leaf node, thereby obtaining the user interaction type that reflects the target user's willingness to interact with the product based on the leaf node.

[0067] For example, assuming the target feature information is {not a paying user, 2 visits, male, 20 years old}, then the decision tree model can determine the type of user interaction based on this target feature information. Figure 3 As shown, the root node of the decision tree model determines that the target user is not a paying user. The process then flows to the right branch, where the node determines the user's visit count. If the visit count is between 1 and 3, the process moves to the next branch, where the node determines the user's age. Since the target user is 20 years old, they are classified as a "slow user." A slow user is someone with no intention to purchase the target product or a low intention to purchase it. A fast user, on the other hand, is someone with an intention to purchase the target product or a high intention to purchase it.

[0068] In one embodiment, the calculation of the information gain of each user feature in the user feature sample set includes the following steps: the terminal calculates the information entropy of each user feature in the user feature sample set to obtain a first information entropy; after dividing the user feature sample set into subsets according to each user feature in the user feature sample set, the second information entropy corresponding to each subset is calculated; and based on the first information entropy and the second information entropy, the information gain of each user feature in the user feature sample set is determined.

[0069] The user feature sample set is a set of features formed based on user features extracted from user data (including basic data and behavioral data) of multiple sample users.

[0070] For example, suppose there is user data D = {d1, d2, ..., dk} from k sample users. k For each user data d in user data D i Each has a user feature sample set, let this user feature sample set be X = {x1, x2, ..., x...} n} contains n-dimensional user features. Since user interaction types include fast users and slow users, the k sample users can be divided into two different categories, namely, the category set C = {c1, c2}. If c i The probability of occurrence is p iThen the information entropy of classification set C (i.e., the first information entropy mentioned above) is:

[0071]

[0072] Next, the conditional entropy of the user feature sample set is calculated, which is the information entropy of the subsets obtained after partitioning the user feature sample set according to user features (i.e., the second information entropy). The specific calculation method is as follows: assuming that a user feature has n distinct values ​​in the user feature sample set (e.g., if the user feature is gender, then this user feature has 2 distinct values ​​in the user feature sample set, i.e., male and female), then the user feature sample set can be divided into n subsets according to user feature x. The information entropy of the subsets after partitioning according to user feature x is:

[0073]

[0074] Among them, |C i |For subset C i The number of samples, i = 1, 2, ..., n; |C| is the total number of samples.

[0075] After calculating the first information entropy and the second information entropy corresponding to the subset division based on user feature x, the information gain after dividing the user feature sample set according to user feature x can be obtained based on the first and second information entropies, as follows:

[0076] Gain(C,x)=Entropy(C)-Entropy x (C)

[0077] The information gain of each user feature within the user feature sample set can be obtained through the above calculation method. This information gain can be found in Table 1. When constructing a decision tree model, the top n user features with the highest information gain values ​​within the user feature sample set can be selected as the optimal subset of user features.

[0078] Table 1 Information Gain of User Characteristics

[0079]

[0080]

[0081] Table 1 can be used to select the top 4 user features with the highest gain values. The optimal user feature sample subset S = {whether the user is a paying user, the number of times the user has viewed the product in the past 7 days, the user's age, and the user's gender}.

[0082] S208 determines the tasks to be participated in and the order in which each task is displayed based on the user interaction type; the rewards corresponding to the tasks are used to interact with the target product.

[0083] The tasks to be participated in can be activities that target users can engage in, such as browsing insurance products, joining a health plan, joining a good car owner plan, receiving free insurance, adding the concierge's WeChat, scheduling a live stream, participating in a lucky draw, and inviting friends. Different user interaction types can have different task displays and order. For example, the task display order for fast users would be: browsing insurance products → joining a health plan → joining a good car owner plan → receiving free insurance → adding the concierge's WeChat → scheduling a live stream → participating in a lucky draw → inviting friends. The task display order for slow users would be: joining a health plan → joining a good car owner plan → receiving free insurance → participating in a lucky draw → scheduling a live stream → browsing insurance products → joining the community → adding the concierge's WeChat → inviting friends.

[0084] The task display order can refer to the order in which the tasks are displayed on the display page. For example, for quick users, the task of browsing insurance products can be displayed in the first task area of ​​the display page, then the task of joining a health plan can be displayed in the second task area, and so on, with the task of inviting friends displayed in the last task area.

[0085] Different tasks can correspond to different rewards. That is, after a target user participates in task A, they can receive corresponding rewards based on the completion of the task, such as coupons, which can be used to purchase corresponding products.

[0086] In one embodiment, the terminal can determine the task display order based on the target user's target feature information, and this task display order can be referenced. Figure 4 .

[0087] S210 generates a display page that matches the user interaction type based on the task, the task display order, and the target product.

[0088] The target product can be related to the product type associated with the triggered entry control. For example, if the entry control is related to insurance, clicking the entry control will take you to a display page for insurance products (i.e., insurance products). Therefore, the target product is an insurance product.

[0089] In one embodiment, page structure data is obtained, and the page structure data is rendered according to the tasks, the task display order, and the target product to obtain a display page that matches the user interaction type. This display page shows the tasks and the target product in the order they are displayed.

[0090] In one embodiment, S206 may specifically include: determining the interaction willingness based on the target feature information using a decision tree model to obtain the target user's degree of willingness to interact with the product; determining the user interaction type based on the degree of willingness, wherein the process of determining the interaction willingness can refer to... Figure 3User interaction types include a first interaction type where the degree of willingness to interact with the product is greater than a first threshold, and a second interaction type where the degree of willingness is less than a second threshold. The first and second thresholds can be the same or different. Therefore, S210 can be divided into the following two scenarios for explanation:

[0091] Scenario 1: For users with the first type of interaction, display the first display page.

[0092] In one embodiment, when the user interaction type is a first interaction type, the terminal generates a first display page matching the first user interaction type based on the task, the task display order, and the target product, and then displays the first display page, such as... Figure 5 As shown.

[0093] Among them, users of the first interaction type can be called fast users, that is, users who are willing to buy the target product or have a high degree of willingness to buy the target product.

[0094] Specifically, the decision tree model is constructed based on the optimal user feature sample subset S = {whether the user is a paying user, number of product views in the past 7 days, user age, user gender} obtained above, and the corresponding values ​​of the user feature sample subset. For example, the value for paying users can be divided into "yes" and "no". When determining user interaction intention, the target feature information corresponding to the user feature sample subset is extracted from the user information, and the target feature information of the target user is judged based on the decision tree model to obtain the user interaction type. For example, if the user's interaction intention level is greater than 0.5, they are considered a fast user; if it is less than 0.5, they are considered a slow user. By dividing and displaying different pages based on the user's target feature information through the decision model, the computational load is small, ensuring the user's page loading speed while improving the accuracy of page display.

[0095] In one embodiment, the first display page is a short version page; the method further includes: when the user interaction type is a first interaction type, the terminal displays theme information in the theme display area of ​​the short version page, displays a preset number of tasks matching the first interaction type in the task display area of ​​the short version page, and hides the remaining tasks matching the first interaction type; and displays a target product in the product display area of ​​the short version page, the target product may include products with normal selling prices and products sold at discounted prices during discount activities, such as... Figure 5 and Figure 7 The short version of the page features a theme display area, a task display area, and a product display area all located within the screen. This effectively improves page display efficiency and facilitates quick user interaction with products, such as enabling quick users to purchase insurance products.

[0096] The short version page refers to a smaller page for the theme display area and task display area. It can display the entire theme display area, task display area, and product display area within the current screen. This allows users with a high intention to purchase products to see the short version page. The tasks on this short version page are displayed in a collapsed manner, with most of the area showcasing insurance products, such as... Figure 7 As shown.

[0097] The task display order configured in the short version page is different from that configured in the long version page. For example, in the task display area of ​​the short version page, a preset number of tasks are displayed according to the task display order corresponding to the first interaction type, and tasks exceeding the preset number are hidden; and / or, at least one of the tasks configured in the long version page is different from the tasks configured in the long version page. For example, task a is displayed in the task display area of ​​the short version page, but not in the task display area of ​​the long version page.

[0098] Furthermore, corresponding templates are pre-defined for both short and long versions of the page. These templates include a theme display area, a task display area, and a product display area, and may also include rendering information, background element design information, and product display parameters. The theme information to be displayed is determined based on the current time. The theme information for the short version page can be modified according to different time periods, including the time schedule for each task, theme information, and rewards earned from participating in past tasks. For example, during the Chinese New Year period, the theme might be railway passenger accident insurance, while for International Women's Day, it might be women's cancer prevention insurance. Besides displaying based on time points, the tasks displayed in the task display area can be popular tasks extracted from the task library with high user participation, such as tasks with a high number of participants (e.g., tasks reaching a certain threshold for popularity). In the short version page, the theme display area can show the time schedule for each task, such as "Draw XX prize on date 1," where "XX prize" refers to the prize that the target user can obtain by participating in a task (such as a lottery) on date 1.

[0099] When a short version page is generated, the template and related information are retrieved, such as theme information, rendering information, background element design information, and product display parameters. Tasks are also retrieved and sorted in the task display area, and the page is rendered to obtain the short version page. Tasks are sorted and displayed based on the user's target characteristics, and themes are displayed according to different time points or popular activities to improve the accuracy of page display.

[0100] Furthermore, determining the product display parameters includes: filtering out product display parameters that match the target user from all available parameters. Specifically, based on the target user's basic and behavioral data, the system identifies products the user has already purchased and uses the missing products as the target user's product display parameters; if the target user already has all the necessary products, it recommends products needed by other family members as the target user's product display parameters.

[0101] In one embodiment, for a hidden task, the terminal can display the hidden task in the task display area according to the input display operation; in addition, the terminal can also display the hidden task in the task display area according to the input swipe operation, and the previously displayed task can be moved to the top, bottom, left or right of the display screen according to the swipe direction.

[0102] In one embodiment, when a user-input task trigger operation is detected, the user can be directed to the task details page of the triggered target task and execute the target task, such as joining a health plan, on that page.

[0103] Scenario 2: For users with the second type of interaction, the first display page is shown.

[0104] In one embodiment, when the user interaction type is the second interaction type, the terminal generates a second display page matching the second user interaction type based on the task, the task display order, and the target product, such as... Figure 6 As shown.

[0105] Among them, users of the second interaction type can be called slow users, that is, users who have no intention to buy the target product or have a low degree of intention to buy the target product.

[0106] In another embodiment, the second display page is a long page; the method further includes: when the user interaction type is the second interaction type, displaying theme information in the theme display area of ​​the long page, the theme information of the long page including the time arrangement information of each task and resource sharing reward activities; displaying all tasks matching the second interaction type in the task display area of ​​the long page, the theme display area and the task display area of ​​the long page being located within the display screen; when the user swipes the page on the display screen and the product display area of ​​the long page appears on the display screen due to the swipe, displaying the target product in the product display area of ​​the long page;

[0107] The main theme display area on the long page is larger than that on the short page, and the task display area on the long page is also larger. This means that the main theme and task display areas on the long page are larger, so only the entire main theme and task display areas are shown on the current screen, while the product display area is displayed when scrolling. This allows slower users with low purchase intent to see the long page, where the tasks and gameplay elements are fully showcased. Figure 8 As shown, slow users can be rewarded and educated by participating in various tasks (short videos to popularize insurance knowledge, insurance product courses), which encourages them to interact with the product using these rewards, thereby helping to increase the purchase rate of insurance products.

[0108] Furthermore, the theme information and the short version of the page can be treated identically. That is, the theme can be displayed based on different time points or popular activities. To increase user awareness of the interactive tasks, the background element design information in the short and long version theme display areas can be partially identical. If the theme information is the same, the promotional activity rendering can be enhanced by expanding the theme information display area in the template. The activity atmosphere rendering can be adjusted and expanded accordingly through the background element design information. When it is a long version page, the template and corresponding information for the long version page are obtained, such as theme information, rendering information, background element design information, product display parameters, etc., and the tasks already sorted in the task display area are retrieved. The page is then rendered to obtain the long version page.

[0109] In addition, the task display order configured in the long version page is different from that configured in the short version page. For example, in the task display area of ​​the long version page, all tasks are displayed in the order corresponding to the second interaction type.

[0110] In addition, the display styles of the first and second display pages can also be different. For example, the display style of the first display page can be relatively simple, while the display style of the second display page can be relatively rich.

[0111] In one embodiment, if at least two target tasks are configured on the display page; after S210, the method further includes: the terminal responding to the task processing operation and sequentially entering the task page of each target task; when the corresponding target task is completed on the task page, obtaining the reward of the corresponding target task; and interacting with products on the display page based on the reward, such as purchasing the corresponding insurance product.

[0112] In the above embodiments, feature extraction is performed on basic data and behavioral data. A decision tree model is used to determine the interaction willingness based on the extracted target feature information, resulting in user interaction types that reflect the target user's willingness to interact with the product. Based on the user interaction type, the tasks to be participated in and the task display order are determined. According to the tasks, task display order, and target product, a display page matching the user interaction type is generated. This allows for different display pages to be generated for users with different interaction types, and different tasks and task display orders are set for different users in a targeted manner. This enables users to participate in corresponding tasks and receive rewards to interact with the product, thus improving the product's interaction rate. Furthermore, the decision tree model is constructed based on user features whose information gain meets the modeling conditions within the user feature sample set. This makes the constructed decision tree model more accurate in judging user interaction types, resulting in more precise display pages shown to users. This effectively encourages users to participate in corresponding tasks and receive rewards to interact with the product, further improving the product's interaction rate.

[0113] In one embodiment, such as Figure 6 As shown, the specific steps for constructing a decision tree model may include:

[0114] S602, Obtain user feature sample set.

[0115] The user feature sample set is a collection of user features from different sample users.

[0116] In one embodiment, the terminal acquires basic data and behavioral data of sample users, and then extracts user features from the basic data and behavioral data to obtain a user feature sample set.

[0117] S604: Select user features whose information gain meets the modeling conditions from the user feature sample set to obtain a subset of user feature samples composed of user features whose information gain meets the modeling conditions.

[0118] Among them, the modeling condition can be a gain threshold. Therefore, user features with information gain not less than the gain threshold are selected from the user feature sample set. The selected user features form a user feature sample subset, which is the optimal feature subset for constructing the decision tree model.

[0119] In one embodiment, S604 may specifically include: the terminal sorting user features in the feature sample set in descending order according to the magnitude of information gain; selecting the user features ranked first in the sorted feature sample set; and combining the selected user features into a subset of user feature samples.

[0120] S606: Extract the user feature with the largest information gain from the subset of user feature samples and use it as the root node.

[0121] S608 uses the root node as a condition to judge user attributes or user behavior and obtains the target branch.

[0122] S610: Within the subset of user feature samples, select the user feature with the second largest information gain as the node of the target branch, and continue until the leaf node is reached to complete the construction of the decision tree model.

[0123] For example, as shown in Table 1, an optimal user feature sample subset S = {whether the user is a paying user, number of times the user viewed the product within 7 days, user age, user gender} can be selected. The root node is chosen based on the information gain of each user feature within the user feature sample subset S, i.e., the user feature with the largest information gain is selected as the root node. For example, paying users can be selected as the root node. Then, the user feature sample subset is divided into two subsets, D1 and D2, based on the root node, where D1 represents the subset corresponding to paying users and D2 represents the subset corresponding to non-paying users. Next, the information gain of the remaining user features in the above subsets is calculated, and the node with the largest information gain is selected as the branch node. For example, for the non-paying branch, the information gain of the remaining user features {number of times the user viewed the product within 7 days, user age, user gender} in the D2 subset is calculated, and "number of times the user viewed the product within 7 days" with the largest information gain is selected as the node for this branch. This process is iterated until the decision numbers are constructed. (See reference...) Figure 3 .

[0124] In the above embodiments, the decision tree model is constructed based on user features whose information gain in the user feature sample set meets the modeling conditions. This makes the constructed decision tree model more accurate in judging the type of user interaction, and thus the display page shown to the user is more precise. This can effectively encourage users to participate in corresponding tasks to get rewards and interact with the product, which is conducive to improving the product's interaction rate.

[0125] It should be understood that, although Figure 2 , 9 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 , 9 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0126] In one embodiment, such as Figure 10As shown, a display page generation device is provided. This device can be a software module, a hardware module, or a combination of both, integrated into a computer device. Specifically, the device includes: an acquisition module 1002, an extraction module 1004, a decision module 1006, a determination module 1008, and a generation module 1010, wherein:

[0127] The acquisition module 1002 is used to acquire the target user's basic data and behavioral data in response to page entry trigger operations;

[0128] Extraction module 1004 is used to extract features from basic data and behavioral data to obtain target feature information;

[0129] Decision module 1006 is used to make decisions on interaction willingness based on target feature information based on decision tree model, and obtain user interaction type that reflects the target user's willingness to interact with the product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions;

[0130] The determination module 1008 is used to determine the tasks to be participated in and the task display order based on the user interaction type; the rewards corresponding to the tasks are used to interact with the target product.

[0131] The generation module 1010 is used to generate a display page that matches the user interaction type based on the task, the task display order, and the target product.

[0132] In the above embodiments, feature extraction is performed on basic data and behavioral data. A decision tree model is used to determine the interaction willingness based on the extracted target feature information, resulting in user interaction types that reflect the target user's willingness to interact with the product. Based on the user interaction type, the tasks to be participated in and the task display order are determined. According to the tasks, task display order, and target product, a display page matching the user interaction type is generated. This allows for different display pages to be generated for users with different interaction types, and different tasks and task display orders are set for different users in a targeted manner. This enables users to participate in corresponding tasks and receive rewards to interact with the product, thus improving the product's interaction rate. Furthermore, the decision tree model is constructed based on user features whose information gain meets the modeling conditions within the user feature sample set. This makes the constructed decision tree model more accurate in judging user interaction types, resulting in more precise display pages shown to users. This effectively encourages users to participate in corresponding tasks and receive rewards to interact with the product, further improving the product's interaction rate.

[0133] In one embodiment, such as Figure 11 As shown, the device also includes:

[0134] The calculation module 1012 is used to calculate the information entropy of each user feature in the user feature sample set to obtain the first information entropy; after dividing the user feature sample set into subsets according to each user feature in the user feature sample set, the second information entropy corresponding to each subset is calculated; based on the first information entropy and the second information entropy, the information gain of each user feature in the user feature sample set is determined.

[0135] In one embodiment, such as Figure 11 As shown, the device also includes:

[0136] Module 1014 is used to acquire a user feature sample set; within the user feature sample set, user features whose information gain meets the modeling conditions are selected to obtain a subset of user feature samples composed of user features whose information gain meets the modeling conditions; the user feature with the largest information gain is taken from the user feature sample subset as the root node; the user attributes or user behaviors are judged based on the root node to obtain the target branch; within the user feature sample subset, the user feature with the second largest information gain is selected as the node of the target branch, until the leaf node is reached to complete the construction of the decision tree model.

[0137] In one embodiment, the construction module 1014 is further configured to sort the user features in the feature sample set in descending order according to the magnitude of information gain; select the user features ranked first in the sorted feature sample set; and combine the selected user features into a subset of user feature samples.

[0138] In the above embodiments, the decision tree model is constructed based on user features whose information gain in the user feature sample set meets the modeling conditions. This makes the constructed decision tree model more accurate in judging the type of user interaction, and thus the display page shown to the user is more precise. This can effectively encourage users to participate in corresponding tasks to get rewards and interact with the product, which is conducive to improving the product's interaction rate.

[0139] In one embodiment, the decision module 1006 is further configured to determine the interaction willingness of the target feature information based on the decision tree model, thereby obtaining the degree of willingness of the target user to interact with the product; and determine the user interaction type according to the degree of willingness; the user interaction type includes a first interaction type with a degree of willingness to interact with the product greater than a first degree threshold and a second interaction type with a degree of willingness less than a second degree threshold;

[0140] like Figure 11 As shown, the device also includes:

[0141] The display module 1016 is used to generate a first display page matching the first user interaction type based on the task, the task display order, and the target product when the user interaction type is the first interaction type; and to generate a second display page matching the second user interaction type based on the task, the task display order, and the target product when the user interaction type is the second interaction type; wherein the task display order configured in the first display page is different from the task display order configured in the second display page; and / or, at least one of the tasks configured in the first display page and the tasks configured in the second display page is different.

[0142] In one embodiment, the first display page is a short page, and the second display page is a long page; the display module 1016 is further configured to: display theme information in the theme display area of ​​the short page when the user interaction type is the first interaction type; display a preset number of tasks matching the first interaction type in the task display area of ​​the short page, and hide the remaining tasks matching the first interaction type; and display the target product in the product display area of ​​the short page; both the task display area and the product display area of ​​the short page are located within the display screen; when the user interaction type is the second interaction type, display theme information in the theme display area of ​​the long page; display all tasks matching the second interaction type in the task display area of ​​the long page, which is located within the display screen; and display the target product in the product display area of ​​the long page when the user swipes the page on the display screen and the product display area of ​​the long page appears on the display screen due to the swipe; wherein the area of ​​the theme display area and the task display area of ​​the short page are both smaller than the area of ​​the theme display area and the task display area of ​​the long page.

[0143] In one embodiment, at least two target tasks are configured in the display page; such as Figure 11 As shown, the device also includes:

[0144] The processing module 1018 is used to respond to task processing operations and sequentially enter the task pages of each target task; when the corresponding target task is completed on the task page, the reward for the corresponding target task is obtained; and the product in the display page is interacted with based on the reward.

[0145] Specific limitations regarding the display page generation device can be found in the limitations of the display page generation method described above, and will not be repeated here. Each module in the aforementioned display page generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0146] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 12 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for generating a display page. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0147] Those skilled in the art will understand that Figure 12 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.

[0148] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0149] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0150] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the computer device to perform the steps in the above method embodiments.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for generating a display page, characterized in that, The method includes: In response to actions triggered by page entry points, obtain basic and behavioral data of the target user; The target feature information is obtained by extracting features from the basic data and the behavioral data using a feature extraction tool; the target feature information is determined based on the extraction parameters configured in the feature extraction tool. The decision tree model is used to make a decision on the interaction willingness of the target feature information to obtain the user interaction type that reflects the target user's interaction willingness with the target product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions; the target product is related to the product type associated with the entry control, and the entry control is the control triggered by the page entry trigger operation in the product entry page; Based on the user interaction type, the tasks to be participated in and the task display order of each task are determined; the task display order is different for different user interaction types; the tasks to be participated in are activities that the target user can participate in; Based on the task, the task display order, and the target product, a display page matching the user interaction type is generated. The display page can be a short page or a long page, and the area of ​​the task display area of ​​the short page is smaller than the area of ​​the task display area of ​​the long page.

2. The method according to claim 1, characterized in that, The method further includes: The information entropy of each user feature in the user feature sample set is calculated to obtain the first information entropy; after dividing the user feature sample set into subsets according to each user feature in the user feature sample set, the second information entropy corresponding to each subset is calculated. Based on the first information entropy and the second information entropy, the information gain of each user feature in the user feature sample set is determined.

3. The method according to claim 1, characterized in that, The model construction based on user features whose information gain in the user feature sample set meets the modeling conditions includes: Obtain a user feature sample set; Within the user feature sample set, user features whose information gain meets the modeling conditions are selected to obtain a subset of user feature samples composed of user features whose information gain meets the modeling conditions. Extract the user feature with the largest information gain from the subset of user feature samples and use it as the root node; The target branch is obtained by judging user attributes or user behavior based on the root node as the judgment condition; Within the subset of user feature samples, the user feature with the second largest information gain is selected as the node of the target branch, and the decision tree model is completed when the leaf node is reached.

4. The method according to claim 3, characterized in that, The step of selecting user features from the user feature sample set whose information gain meets the modeling condition, to obtain a subset of user feature samples composed of user features whose information gain meets the modeling condition, includes: The user features in the feature sample set are sorted in descending order according to the magnitude of the information gain. Within the sorted feature sample set, select the user features ranked first by a preset order; The selected user features are combined into a subset of user feature samples.

5. The method according to any one of claims 1 to 4, characterized in that, The decision tree model-based approach to determine the interaction intention based on the target feature information yields user interaction types that reflect the target user's willingness to interact with the product, including: The interaction intention is determined based on the target feature information using a decision tree model to obtain the degree of willingness of the target user to interact with the product; the user interaction type is determined based on the degree of willingness; the user interaction type includes a first interaction type with a degree of willingness to interact with the product greater than a first degree threshold and a second interaction type with a degree of willingness less than a second degree threshold; The step of generating a display page matching the user interaction type based on the task, the task display order, and the target product includes: When the user interaction type is the first interaction type, a first display page matching the first interaction type is generated based on the task, the task display order, and the target product. When the user interaction type is the second interaction type, a second display page matching the second interaction type is generated based on the task, the task display order, and the target product. Wherein, the task display order configured in the first display page is different from the task display order configured in the second display page; and / or, at least one of the tasks configured in the first display page is different from the tasks configured in the second display page.

6. The method according to claim 5, characterized in that, The first display page is a short version, and the second display page is a long version; the method further includes: When the user interaction type is the first interaction type, theme information is displayed in the theme display area of ​​the short version page, a preset number of tasks matching the first interaction type are displayed in the task display area of ​​the short version page, and the remaining tasks matching the first interaction type are hidden; and the target product is displayed in the product display area of ​​the short version page; the theme display area, task display area and product display area of ​​the short version page are all located within the display screen; When the user interaction type is the second interaction type, theme information is displayed in the theme display area of ​​the long page; all tasks matching the second interaction type are displayed in the task display area of ​​the long page, and the theme display area and task display area of ​​the long page are located within the display screen; when the page is swiped on the display screen and the product display area of ​​the long page appears in the display screen due to swiping, the target product is displayed in the product display area of ​​the long page. The area of ​​the theme display area on the short version page is smaller than the area of ​​the theme display area on the long version page.

7. The method according to any one of claims 1 to 4, characterized in that, The display page is configured with at least two target tasks; After generating the display page that matches the user interaction type, the method further includes: In response to the task processing operation, the task pages of each target task are entered sequentially; When the corresponding target task is completed on the task page, the reward for the corresponding target task will be obtained; Interact with the products on the display page based on the rewards.

8. A display page generation device, characterized in that, The device includes: The acquisition module is used to acquire basic and behavioral data of the target user in response to page entry trigger operations; The extraction module is used to extract features from the basic data and the behavioral data using a feature extraction tool to obtain target feature information; the target feature information is determined based on the extraction parameters configured in the feature extraction tool. The decision module is used to make decisions on the interaction willingness based on the target feature information based on the decision tree model, and to obtain the user interaction type that reflects the target user's interaction willingness with the target product; the decision tree model is obtained by constructing a model based on user features in the user feature sample set whose information gain reaches the modeling conditions; the target product is related to the product type associated with the entry control, and the entry control is the control triggered by the page entry trigger operation in the product entry page; The determining module is used to determine the task to be participated in and the task display order of each task based on the user interaction type; the task display order is different for different user interaction types; the task to be participated in is an activity that the target user can participate in; The generation module is used to generate a display page that matches the user interaction type based on the task, the task display order, and the target product. The display page can be a short page or a long page, and the area of ​​the task display area of ​​the short page is smaller than the area of ​​the task display area of ​​the long page.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images