Page display method, device, medium and electronic device based on user characteristics

Abstract

The present invention provides a page display method, device, medium and electronic device based on user characteristics. When it is necessary to recommend a product to a certain user, label data representing the characteristics of the user is obtained, and the similarity between the user and each of a plurality of user groups obtained in advance is calculated based on the label data. Then, based on the similarity, one or more target user groups with the highest similarity to the user are selected from the plurality of user groups, and then page modules corresponding to these target user groups are obtained. Then, on a recommendation page displayed to the user, product information is recommended through these page modules respectively. Products of one attribute are clustered in each page module. By considering the characteristics of the user, products corresponding to the user group most similar to the user are recommended to the user, which is closer to the user's preferences, and product recommendations can be made dynamically for different users without the need to produce material pictures separately, which is simple to implement.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis technology, and in particular to a page display method, device, medium and electronic device based on user characteristics. Background Art

[0002] With the rapid development of Internet technology, users buy products through online shopping, which has become a daily routine. There are a large number of products on shopping websites, and it is a waste of time for users to screen them. Therefore, many shopping websites or online shopping applications recommend products.

[0003] At present, product recommendations are mainly made through landing pages. Most landing pages rely on landing page tools to be manually built, and material pictures are made, components are selected, product skus are uploaded, and page structure is adjusted to complete page construction and obtain the corresponding landing page. After the user opens the web page or application, the landing page is displayed, or the landing page is pushed to users who have subscribed to the recommendation to recommend products to the user.

[0004] However, the current product recommendation method is complex to implement and has high implementation costs. In addition, different users require different products. Users still need a lot of time to screen products using the recommendation method in the above solution. Summary of the invention

[0005] The embodiments of the present invention provide a page display method, device, medium and electronic device based on user characteristics to solve the problems in the prior art that the product recommendation method is complex to implement, has high implementation costs, and requires users to spend more time to screen products.

[0006] In a first aspect, an embodiment of the present invention provides a page display method based on user characteristics, comprising:

[0007] Acquire tag data of a user, where the tag data is used to represent features of the user;

[0008] Calculating the similarity between the user and each pre-acquired user group according to the tag data; each user group is used to represent a type of user;

[0009] Acquire at least one target user group according to the similarity between the user and each user group, wherein the similarity between the at least one target user group and the user is greater than the similarity between other user groups and the user;

[0010] Acquire at least one page module corresponding to the at least one target user group;

[0011] The at least one page module is displayed on a recommendation page corresponding to the user, and each page module includes information of at least one product.

[0012] In a specific implementation, the recommendation page also displays clustering attributes of the products in each page module.

[0013] Optionally, the clustering attributes include at least: selected products, hot-selling products, well-received products, or newly launched products.

[0014] In a specific implementation, calculating the similarity between the user and each pre-acquired user group according to the tag data includes:

[0015] For each user group, a cosine similarity calculation method is used to calculate the cosine similarity between the tag data and the feature information of the user group, and the similarity between the user and the user group includes the cosine similarity.

[0016] In a specific implementation, the method of calculating the cosine similarity between the tag data and the feature information of the user group by using a cosine similarity calculation method includes:

[0017] Using the cosine similarity formula:

[0018]

[0019] Calculate and obtain the cosine similarity between the label data and the characteristic information of the user group; where u j Indicates characteristic information of the user group, represents the label data of the user A, and j represents the number of features.

[0020] In a specific implementation, acquiring at least one target user group according to the similarity between the user and each user group includes:

[0021] Sort the similarities between the user and each user group from large to small or from small to large, and obtain at least one target similarity with the largest similarity starting from the largest similarity;

[0022] At least one corresponding target user group is acquired according to the at least one target similarity.

[0023] In a specific implementation, the number of target user groups is determined based on the recommendation pages.

[0024] In a specific implementation, the step of obtaining the user's tag data includes:

[0025] Upon receiving a page acquisition request carrying the identity information of the user, acquiring the historical consumption data of the user according to the identity information of the user;

[0026] The tag data of the user is obtained according to the historical consumption data.

[0027] In a specific implementation, the step of obtaining the user's tag data includes:

[0028] Obtaining historical consumption data of a user according to the identity information of the user who subscribes to the page recommendation in the page recommendation subscription information;

[0029] The tag data of the user is obtained according to the historical consumption data.

[0030] In a specific embodiment, the method further comprises:

[0031] According to the channel information accessed by the user, the recommendation page corresponding to the channel information is obtained, and at least one page module can be displayed on the recommendation page.

[0032] In another specific embodiment, the method further comprises:

[0033] Obtaining a label feature vector for each user according to the annotated user data, wherein the label feature vector includes multiple labels corresponding to the user;

[0034] Cluster users according to the feature vector of each user to obtain multiple user groups;

[0035] For each user group, the CBOW model is used according to the feature vector of each user in the user group to calculate the feature information corresponding to the user group.

[0036] In another specific embodiment, the method further comprises:

[0037] Configure page modules according to clustering attributes of multiple products to obtain multiple page modules;

[0038] According to the characteristic information of each user group, a mapping relationship between the page module and the user group is established and stored.

[0039] In yet another specific implementation, the acquiring at least one page module corresponding to the at least one target user group includes:

[0040] At least one page module corresponding to the at least one target user group is acquired according to the mapping relationship between the page modules and the user groups.

[0041] In a second aspect, an embodiment of the present invention provides a page display device based on user characteristics, including:

[0042] An acquisition module, used to acquire tag data of a user, wherein the tag data is used to represent features of the user;

[0043] A processing module, used for calculating the similarity between the user and each pre-acquired user group according to the tag data; each user group is used to represent a type of user;

[0044] The processing module is further used to obtain at least one target user group according to the similarity between the user and each user group, wherein the similarity between the at least one target user group and the user is greater than the similarity between other user groups and the user;

[0045] The processing module is further used to obtain at least one page module corresponding to the at least one target user group;

[0046] An interactive module is used to display the at least one page module on the recommendation page corresponding to the user, and each page module includes information of at least one product.

[0047] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0048] processor, memory, and,interface;

[0049] A memory, configured to store executable instructions of the processor;

[0050] Wherein, the processor is configured to execute the page display method based on user features as described in any one of the first aspects by executing the executable instructions.

[0051] In a fourth aspect, an embodiment of the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the page display method based on user features as described in any one of the first aspects.

[0052] The user feature-based page display method, device, medium and electronic device provided by the embodiments of the present invention obtain label data representing the features of the user when it is necessary to recommend a product to a certain user, calculate the similarity between the user and each of the multiple user groups obtained in advance based on the label data, and then select one or more target user groups with the highest similarity to the user among the multiple user groups based on the similarity, and then obtain page modules corresponding to these target user groups, and then recommend product information through these page modules on the recommendation page displayed to the user. Products of one attribute are clustered in each page module, and products corresponding to the most similar user group are recommended to the user by considering the user's features, which is closer to the user's preferences, and can dynamically make product recommendations for different users without the need to separately produce material images, which is simple to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0054] Figure 1 A flowchart of a first embodiment of a method for displaying a page based on user characteristics provided in an embodiment of the present application;

[0055] Figure 2 A schematic diagram of a recommendation page provided in an embodiment of the present application;

[0056] Figure 3 A schematic diagram of the overall architecture of the page provided for this application;

[0057] Figure 4 A flowchart of a second embodiment of a method for displaying a page based on user characteristics provided in an embodiment of the present application;

[0058] Figure 5 A schematic diagram of a One-Hot representation of a user's label feature vector provided in an embodiment of the present application;

[0059] Figure 6 A schematic diagram of word2vec characterization of user features provided in an embodiment of the present application;

[0060] Figure 7 A flowchart of a third embodiment of a method for displaying a page based on user characteristics provided in an embodiment of the present application;

[0061] Figure 8 A schematic diagram of the structure of a first embodiment of a page display device based on user characteristics provided in an embodiment of the present application;

[0062] Fig. 9 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiment of the present invention, all other embodiments made by ordinary technicians in this field under the enlightenment of this embodiment belong to the scope of protection of the present invention.

[0064] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0065] With the rapid development of Internet technology, the scope of advertising business has gradually expanded, and advertising landing pages have attracted more and more attention. Advertising landing pages, that is, pages displayed to users, can also be called recommendation pages. In the existing technology, it is necessary to manually build landing pages, make material pictures, select components, upload products, and adjust the page structure to complete the page construction and obtain the corresponding landing page.

[0066] However, this product recommendation method is complex to implement and has high implementation costs. In addition, different users require different products. Users still need a lot of time to screen products using the recommendation method in the above solution.

[0067] An advertising landing page usually refers to the page that users jump to after clicking on a web page or video. The purpose of a landing page is to recommend suitable products to users, facilitate their purchase of products, and improve the efficiency of product transactions on the seller side. With the continuous rise of big data, in order to reduce the cost of advertisers making landing pages, smart landing pages have gradually come into everyone's view. Each smart landing page sets the module for page display, and then classifies the products (or products) in the store by different dimensions. For example, products are selected from dimensions such as sales volume, good reviews, and new products, and filled into the corresponding modules. That is, the products on each floor of the page are aggregated according to the floor attributes, and then sorted according to the estimated click-through rate to form a complete smart landing page. The front-end page rendering and theme color matching are fixed, and the page hierarchy and module order are also fixed.

[0068] In view of the above situation, the present application provides a page display method based on user characteristics. The core idea of ​​the solution is a technical solution for dynamic display of intelligent landing pages based on user tag data. It utilizes known user preferences for page modules and acquired user characteristics, and can also be based on some important factors of media channel information to complete the dynamic display of the module level of the recommended page to the user, so that the recommended page obtained by each user is no longer the same and without novelty, while achieving the effect of recommending products to users. In addition, the use of this intelligent method has a simple implementation process and low cost, and can also solve the problem that it is difficult to conduct comparative experiments on recommended pages.

[0069] Below, the page display solution based on user features provided by the present application is described in detail through several specific implementations.

[0070] This solution can be applied in shopping websites or merchants' servers or other service platforms, or can also be applied in cloud platforms that can obtain user account information and historical data such as historical consumption records, or other electronic devices. This solution does not impose any restrictions on this.

[0071] Figure 1 A flowchart of a first embodiment of a page display method based on user features provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the page display method based on user characteristics specifically includes the following steps:

[0072] S101: Obtaining user tag data, where the tag data is used to represent user features;

[0073] In this step, when product information needs to be recommended to a user, the user's tag data needs to be obtained, that is, some tags that can represent the user's characteristics, which can be tag data obtained by one or more tags. Generally speaking, a tag or feature is a specific word and cannot be directly calculated, so each word can be identified as a numerical word vector to form the tag data here.

[0074] In the specific implementation of this solution, some data of the user can be obtained based on the user's identity information (such as the login account, or cookie information) to perform profiling and obtain the above-mentioned label data, or for users who are not logged in, the device dimension portrait data storage table can be used to obtain the above-mentioned label data (mainly used for off-site users).

[0075] In another specific implementation of the solution, if the solution is applied to a web page or an application (for example, a shopping website, a shopping application, etc.), the server obtains user tag data for page recommendation, including at least the following scenarios: a registered user subscribes to a product recommendation or page recommendation function; or the user authorizes the platform to make product recommendations; or the user makes product recommendations after entering the application, or the user clicks on an advertisement to enter the recommendation page. In all of these scenarios, it is necessary to obtain the user's tag data. The following is an example of how to obtain the user's tag data:

[0076] In the first implementation, after the user clicks on an advertisement or a pushed notification, when the server receives a page acquisition request carrying the user's identity information, the server acquires the user's historical consumption data based on the user's identity information, and acquires the user's tag data based on the historical consumption data. That is, under the user's trigger, the server acquires the historical consumption data based on the user's identity information, and analyzes the historical consumption data to acquire the user's tag data.

[0077] In the second implementation, if the user subscribes to the recommendation information, when the preset recommendation time arrives, the user's historical consumption data can be obtained according to the identity information of the user who subscribes to the page recommendation in the page recommendation subscription information, and the user's tag data can be obtained according to the historical consumption data. Similar to the above, the historical consumption data is obtained according to the user's identity information, so as to analyze and obtain the user's tag data.

[0078] S102: Calculate the similarity between the user and each pre-acquired user group according to the tag data; each user group is used to represent a type of user.

[0079] In this step, after obtaining the user's tag data, it is necessary to determine which user group the user is closer to or belongs to, so as to subsequently select an appropriate page module based on the user group to recommend products to the user.

[0080] In this solution, the similarity between the user and each user group is calculated based on the label data, and the same user group is used to represent a type of user. In the specific calculation, the final similarity value can be calculated based on the feature information of the user group and the label data.

[0081] S103: acquiring at least one target user group according to the similarity between the user and each user group, wherein the similarity between the at least one target user group and the user is greater than the similarity between the other user groups and the user;

[0082] In this step, after calculating the similarity between the user and each user group, it is necessary to obtain one or more user groups that are most similar to the user as the target user group. The specific number of target user groups can be determined according to the number of page modules that can be laid out on the recommendation page, and the specific number is not limited.

[0083] In a specific implementation, the similarities between the user and each user group may be sorted from large to small or small to large, and at least one target similarity with the largest similarity is obtained starting from the one with the largest similarity; and at least one corresponding target user group is obtained according to the at least one target similarity.

[0084] The solution means selecting one or more user groups with the highest similarity to the user from multiple user groups as the target user group.

[0085] S104: Acquire at least one page module corresponding to at least one target user group.

[0086] In this step, after obtaining one or more target user groups, the page modules corresponding to these target user groups can be determined based on the relationship between the predetermined user groups and the page modules. Among them, the page modules are attribute pages associated with the characteristics of the users in the user group. For example, the consumption habits or records of some users can determine that these users like well-received products, and some users like to buy new products, etc. The premise for the implementation of this solution is to classify users into fixed user groups through the analysis of big data, and cluster the consumption history data or historical browsing data of users in each user group, and configure the corresponding page modules to correspond the page modules to the user groups.

[0087] S105: Display at least one page module on a recommendation page corresponding to the user, each page module including information of at least one product.

[0088] In this step, after selecting at least one page module, all of these page modules may be displayed on a recommendation page pushed to the user, and the page modules may include information of one or more products.

[0089] Optionally, in the specific implementation of this solution, it is first necessary to obtain the recommendation page before these page modules can be displayed on the recommendation page. Generally speaking, the framework of the recommendation page can be pre-configured, and it can be directly read when in use, and the corresponding page modules can be laid out in the recommendation page. However, the recommendation page finally presented by the user in the mobile application, or the browser on the PC, or other scenarios is different. Therefore, in the implementation of this step, it is also necessary to obtain the user's access channel, that is, to obtain the user's channel information. Then, according to the channel information accessed by the user, the recommendation page corresponding to the channel information is obtained. It should be understood that the recommendation page layout obtained by different access channels and the number of page modules that can be laid out can be the same or different, and this solution does not limit this.

[0090] When multiple page modules are obtained, these page modules can be laid out on the recommendation page according to preset rules or randomly, and this solution does not impose any restrictions. For example, the page modules corresponding to the target user group with the highest similarity can be laid out first in the order of similarity with the user from high to low, and this solution does not impose any restrictions on this.

[0091] In the specific implementation of this solution, the recommendation pages displayed to different users are different. As the user's consumption habits change, these recommended pages are not static. It is possible to implement a solution that dynamically displays page modules to users.

[0092] Figure 2 A schematic diagram of a recommendation page provided in an embodiment of the present application, such as Figure 2 As shown, in the recommendation page, a total of three page modules are displayed to the user: page module 1, page module 2 and page module 3, and each page module can recommend information about one or more products.

[0093] Optionally, the recommendation page can also display clustering attributes of products in different page modules, for example Figure 2 The selected products, hot-selling products, well-received products, etc. shown in the example. In the specific application of this solution, the clustering attributes at least include: selected products, hot-selling products, well-received products, or newly launched products. Clustering recall can also be performed based on various attributes such as sales volume, discounts, limited-time specials, flash sales, special user exclusives, etc. This solution does not limit this.

[0094] The page display method based on user characteristics provided in the present embodiment obtains label data representing the characteristics of the user when it is necessary to recommend a product to a certain user, calculates the similarity between the user and each of the multiple user groups obtained in advance based on the label data, and then selects one or more target user groups with the highest similarity to the user among the multiple user groups based on the similarity, and then obtains the page modules corresponding to these target user groups, and then recommends product information through these page modules on the recommendation page displayed to the user. Products of one attribute are clustered in each page module, and the product corresponding to the most similar user group is recommended to the user by considering the characteristics of the user, which is closer to the user's preferences, and can dynamically make product recommendations for different users without the need to separately produce material images, which is simple to implement.

[0095] On the basis of the above embodiments, it can be known that in the page display method based on user features provided by this solution, there is a certain mapping relationship between the page module in the recommendation page and the user group (that is, the user), and the user's label data is also called user portrait data, which can be stored in the form of a table. In all embodiments of this application, it should be understood that the words such as label data, user portrait, user features, feature labels, labels, label vectors, or label feature vectors are all used to represent the characteristics of the user, but they are different forms of representation, with similar meanings, and different data forms required in different processing processes. Similarly, for user groups, feature vectors, feature information, feature data, etc. are also different forms used to represent the characteristics of user groups. In different processing processes, appropriate forms can be matched, and this solution does not impose any restrictions. The overall architecture of the solution is explained below.

[0096] Figure 3 The overall architecture diagram of the page provided for this application is as follows: Figure 3 As shown in the figure, in the whole process of this solution, the identity information of the purchasing user who needs to be recommended is first obtained, and then the user tag, that is, the tag data mentioned above, is obtained from the DMP user portrait data storage table. In the next step, the user's tag data and channel characteristics are combined as the data representation of the user, and the similarity with each user group is calculated. Finally, different page modules are dynamically displayed to each user based on the similarity.

[0097] The recommendation page (i.e., smart landing page) displayed to users is composed of multiple product page modules. The products in each page module are recalled based on certain characteristics of the products in the merchant's store, such as sales, discounts, flash sales, good reviews, new products, plus exclusives, monthly product visits, etc. The corresponding page modules are such as best-selling products (recalled according to product sales), limited-time sales (products with flash sales and flash sales), special offers, etc. Each page module has its corresponding multiple front-end interface styles, such as one row and one column, one row and two columns, and different visual designs (colors).

[0098] Each page module has a corresponding group of purchasing users, and the user tags of the corresponding page module can be obtained (sales promotion sensitivity, product discount rate preference, product evaluation sensitivity, flash sale product preference, color preference, user activity, gender, age, purchasing power, education level, etc.). For example, the users who buy the products in the special offer module with one row and two columns are mainly middle-aged female users, while young male users are more inclined to buy the products in the good reviews module.

[0099] Assuming that there are N page modules in the displayed recommendation page, and each page module has M styles, there are a total of N*M candidate sets. Let Nm be the style m of module N, and the corresponding feature set of user group U is B = {b1, b2, b3, ..., b j}, b j Represents a characteristic of a user.

[0100] like Figure 3 As shown, in the specific mapping, the pin (user account) of the ordering user corresponding to each page module is obtained, and associated with the user portrait data storage table of the pin dimension, so as to obtain the characteristic information of the user group of this module. The accuracy and quality evaluation of the label feature information are guaranteed by the DMP platform. Each feature in the user group is based on the user with the largest number of users in the group. For example, if the first feature of the user group is 'not sensitive to promotions', it means that most users in the user group are not sensitive to promotions.

[0101] In the current product recommendation scenario, the inconsistency of traffic in different channels will also affect user preferences. In order to better characterize user characteristics and further improve the probability of recommendation, the channel characteristic factor T can be added based on the user basic tag data provided by DMP. That is, the characteristic information of the user group corresponding to the page module N is B = {b1, b2, b3, ..., b j , T}.

[0102] The specific user data is characterized as follows:

[0103] Figure 4 A flow chart of a second embodiment of a page display method based on user features provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, based on the above embodiment, the page display method based on user features also includes the following steps:

[0104] S201: Obtain a label feature vector of each user according to the labeled user data, where the label feature vector includes multiple labels corresponding to the user.

[0105] S202: Clustering users according to the feature vector of each user to obtain multiple user groups.

[0106] In the above steps, in order to be able to cluster users into user groups, a large amount of labeled data is required, that is, label feature vectors of many users are required for subsequent processing.

[0107] Different users can be distinguished according to the label feature vectors of the marked users, that is, clustered according to the features to obtain multiple user groups. After different user groups are obtained, the features of users in different user groups can be represented by one feature information.

[0108] S203: For each user group, according to the feature vector of each user in the user group, the CBOW model is used to calculate the feature information corresponding to the user group.

[0109] In this step, Figure 5 A schematic diagram of a One-Hot representation of a user's label feature vector provided in an embodiment of the present application is shown in FIG. Figure 5 As shown in the figure, since the user tag or feature is a specific word, it is impossible to directly calculate the similarity. Each word needs to be represented as a numerical word vector. If there are j features, the first feature is represented as feature value b_1 = [1, 0, 0, 0, 0, ..... 0] (One-Hot Vector), and the nth feature is set to 1 at the nth, and each feature is represented as j dimensions. Then the features of the user group U are represented as i*j dimensions.

[0110] Figure 6 A schematic diagram of a word embedding model (word2vec) in a neural network provided in an embodiment of the present application to characterize user features. The above method is used to characterize the features of a user group as a very high-dimensional and sparse vector. In order to m The user label B of the corresponding user group U is converted from a high-dimensional sparse feature vector to a low-dimensional dense feature vector, which can be processed using word2vec in the neural network, such as Figure 6As shown in the figure. Since the data set is small and the word vector of a specific word needs to be output, the Continuous Bag-of-Words (CBOW) model is used. The training input is the word vector corresponding to the context-related words of a certain feature word, and the output is the word vector of this specific word. If the jth label of the user is represented by b j , whose word vector is u j ∈R d , where d is the user label embedding dimension in the low-dimensional space.

[0111] The input of the model is a one-hot vector, that is, for a given label, the weight matrix between the input layer and the hidden layer is represented by a V×N matrix W. Each row of W is a j-dimensional vector v corresponding to the model input W , after transposition, the i-th row of the matrix is

[0112]

[0113] Where C is the number of context features, i.e. w1, w2, …w i is the eigenvalue in the context, v w is the input vector of word w.

[0114] There is a different matrix W′ between the hidden layer and the output layer, and the dimension of the matrix is ​​N×V. After obtaining the hidden layer weight matrix learned by the model training data, each feature u can be calculated j Score:

[0115] u j =v′ wj T h

[0116] v′ wj is the jth column of the weight matrix W′. The loss function is defined as:

[0117]

[0118] The weight update equations for the hidden layer and the output layer.

[0119] Through the above method, we can get the final feature information representing each user group, that is, the feature vector representing the user group, which is a low-dimensional dense feature vector. This solution uses the word2vec model to represent user tags for user features. Different from direct and simple comparison and traditional machine learning methods, it can continuously iterate and update the algorithm. The data representation of the tag makes the module recommendation results more accurate.

[0120] Figure 7A flow chart of a third embodiment of a page display method based on user features provided in an embodiment of the present application is shown in FIG. Figure 7 As shown in the above Figure 4 Based on the embodiment, the page display method based on user characteristics also includes the following steps:

[0121] S204: Perform page module configuration according to clustering attributes of multiple products to obtain multiple page modules.

[0122] In this step, the above scheme obtains characteristic information of different user groups based on the processing of big data. In order to realize dynamic product recommendation for different user groups, it is necessary to configure page modules corresponding to different user groups.

[0123] In the specific implementation of this solution, the historical consumption data of users in each user group over a period of time, or all consumption data, can be obtained, and the main attribute characteristics of the products purchased by users in a user group can be determined by analyzing the products purchased by the users. The attributes of the products that are most likely to be purchased by the user group can be obtained, and the corresponding page module can be configured according to the attributes.

[0124] S205: Establishing and storing a mapping relationship between page modules and user groups according to the characteristic information of each user group.

[0125] In this step, after obtaining the characteristic information of each user group, the server or service platform needs to establish a mapping relationship between the page module and the user group. Specifically, the mapping relationship can be that the page module directly corresponds to the user group and stores the characteristic information of each group. Alternatively, the characteristic information of the page module and the user group can be directly mapped, and this solution does not limit this. In this solution, the page module is mapped to the user group, the user portrait corresponding to the page module is obtained, and the presentation of the page module is disassembled in a more fine-grained manner.

[0126] Based on this embodiment, in step S104 of embodiment one, obtaining at least one page module corresponding to at least one target user group can be specifically implemented as follows: according to the mapping relationship between the page module and the user group, obtaining at least one page module corresponding to the at least one target user group.

[0127] The page display method based on user characteristics provided in the above embodiment obtains different user groups by clustering the analysis of user characteristics, and configures different page modules for different user groups. In the process of product recommendation, based on user characteristics and focusing on the dimensions of page modules, the recommended page is dynamically displayed, which improves the flexibility of the page, catches the user's attention, and improves the page conversion rate. At the same time, channel information is also considered to further improve the accuracy of recommendation.

[0128] Based on the above scheme, a specific scheme for recommending products to users is introduced below.

[0129] When user A accesses the page and initiates a request, the user's cookie information is obtained. For users who are not logged in, the device dimension portrait data storage table is used to obtain the above selected user tags (mainly external users), and for logged in users, user_log_acct (user login account) is used. Similarly, the word vector embedding is performed on the obtained tags, recorded as

[0130] 1. Calculate the similarity between users and user groups

[0131] For each user group, a cosine similarity calculation method is used to calculate the cosine similarity between the tag data and the feature information of the user group, and the similarity between the user and the user group includes the cosine similarity.

[0132] Specifically, in order to obtain a page module that is more suitable for user A, cosine similarity may be used to calculate the difference between user A and each user group.

[0133]

[0134] Among them, u j Indicates characteristic information of the user group, represents the label data of the user A, and j represents the number of features.

[0135] If the angle between them is smaller, it is determined that the two vectors are more similar, that is, user A purchases user group u j Corresponding page module N m The probability of the product is greater.

[0136] 2. Dynamic display of pages

[0137] After calculating the cosine similarity between user A and each user group, they are arranged from high to low. If the page is set to consist of K modules, the modules with probability of top-K are displayed from top to bottom. In this way, different recommended pages can be displayed for each user, thus realizing the dynamic display technology of pages based on user tags and channel characteristics.

[0138] In particular, for users already in the data set, in order to provide them with a better shopping experience, the last product module on the page is a product recommendation module (Guess What You Like), and the products in the module will be personalized based on user tags.

[0139] The page display scheme based on user characteristics provided in the embodiment of the present application aims to design a dynamic display technology for intelligent landing pages based on user tags and channel characteristics, so as to accurately reach different users, provide users with personalized and accurate product recommendations, realize refined page operations, and improve the conversion effect of landing pages. For systems based on e-commerce platforms, landing pages play an important role in the conversion between recommendations and orders, but when there are more and more channels, diverse category structures, and very different traffic quality, intelligent landing pages cannot efficiently match traffic and optimize pages. This solution mines user tags and channel characteristics to achieve a thousand faces for a thousand people from the dimension of page modules, and dynamically displays each product module of the intelligent landing page, so that each segmented traffic and different types of users have corresponding landing pages, which significantly improves the user experience and increases user stickiness. At the same time, it helps merchants better solve the "long tail" problem of products and improves page conversion rates.

[0140] Figure 8 A schematic diagram of a first embodiment of a page display device based on user characteristics provided in an embodiment of the present application is shown in FIG. Figure 8 As shown, the page display device 10 based on user characteristics includes:

[0141] An acquisition module 11 is used to acquire tag data of a user, where the tag data is used to represent features of the user;

[0142] A processing module 12 is used to calculate the similarity between the user and each pre-acquired user group according to the tag data; each user group is used to represent a type of user;

[0143] The processing module 12 is further used to obtain at least one target user group according to the similarity between the user and each user group, wherein the similarity between the at least one target user group and the user is greater than the similarity between other user groups and the user;

[0144] The processing module 12 is further used to obtain at least one page module corresponding to the at least one target user group;

[0145] The interactive module 13 is used to display the at least one page module on the recommendation page corresponding to the user, and each page module includes information of at least one product.

[0146] The page display device based on user characteristics provided in this embodiment is used to execute the technical solution provided by any of the aforementioned method embodiments. Its implementation principle and technical effect are similar. By considering the characteristics of the user, the product corresponding to the most similar user group is recommended to the user, which is closer to the user's preferences. Product recommendations can be made dynamically for different users without the need to produce material pictures separately, and the implementation is simple.

[0147] On the basis of the above-mentioned embodiment, in the specific implementation of the page display device 10 based on user characteristics, the recommendation page also displays the clustering attributes of the products in each page module, wherein the clustering attributes include at least: selected products, hot-selling products, well-received products, or newly launched products.

[0148] Optionally, the processing module 12 is specifically used for:

[0149] For each user group, a cosine similarity calculation method is used to calculate the cosine similarity between the tag data and the feature information of the user group, and the similarity between the user and the user group includes the cosine similarity.

[0150] Optionally, the processing module 12 is specifically used for:

[0151] Using the cosine similarity formula:

[0152]

[0153] Calculate and obtain the cosine similarity between the label data and the characteristic information of the user group; where u j Indicates characteristic information of the user group, represents the label data of the user A, and j represents the number of features.

[0154] Optionally, the processing module 12 is specifically used for:

[0155] Sort the similarities between the user and each user group from large to small or from small to large, and obtain at least one target similarity with the largest similarity starting from the largest similarity;

[0156] At least one corresponding target user group is acquired according to the at least one target similarity.

[0157] Optionally, the number of target user groups is determined based on the recommendation pages.

[0158] Optionally, the acquisition module 11 is specifically used for:

[0159] Upon receiving a page acquisition request carrying the identity information of the user, acquiring the historical consumption data of the user according to the identity information of the user;

[0160] The processing module 12 is further configured to obtain tag data of the user according to the historical consumption data.

[0161] Optionally, the acquisition module 11 is specifically used for:

[0162] Obtaining historical consumption data of a user according to the identity information of the user who subscribes to the page recommendation in the page recommendation subscription information;

[0163] The processing module 12 is further configured to obtain tag data of the user according to the historical consumption data.

[0164] The processing module 12 is also used for:

[0165] According to the channel information accessed by the user, the recommendation page corresponding to the channel information is obtained, and at least one page module can be displayed on the recommendation page.

[0166] Based on any of the above embodiments, the processing module 12 is further used for:

[0167] Obtaining a label feature vector for each user according to the annotated user data, wherein the label feature vector includes multiple labels corresponding to the user;

[0168] Cluster users according to the feature vector of each user to obtain multiple user groups;

[0169] For each user group, the CBOW model is used according to the feature vector of each user in the user group to calculate the feature information corresponding to the user group.

[0170] Optionally, the processing module 12 is further used for:

[0171] Configure page modules according to clustering attributes of multiple products to obtain multiple page modules;

[0172] According to the characteristic information of each user group, a mapping relationship between the page module and the user group is established and stored.

[0173] Optionally, the processing module 12 is specifically used for:

[0174] At least one page module corresponding to the at least one target user group is acquired according to the mapping relationship between the page modules and the user groups.

[0175] The page display based on user features provided in any of the above embodiments is used to implement the technical solution of any of the above method embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0176] Fig. 9 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Fig. 9 As shown, an electronic device 20 provided in this embodiment includes:

[0177] Processor 21, memory 22, and interface 23;

[0178] A memory 22, used for storing executable instructions of the processor, and the memory 22 may also be a flash memory;

[0179] The processor 21 is configured to execute the various steps in the above method by executing the executable instructions. For details, please refer to the relevant description in the above method embodiment.

[0180] Optionally, the memory 22 may be independent or integrated with the processor 21 .

[0181] When the memory 22 is a device independent of the processor 21, the electronic device may further include:

[0182] The bus 24 is used to connect the processor 21, the memory 22 and the interface 23. The interface 23 includes a communication interface for data transmission and a display interface or an operation interface for human-computer interaction.

[0183] This embodiment also provides a readable storage medium, in which a computer program is stored. When at least one processor of an electronic device executes the computer program, the electronic device executes the page display method based on user features provided in the above-mentioned various implementation modes.

[0184] This embodiment also provides a program product, which includes a computer program stored in a readable storage medium. At least one processor of the electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device implements the page display method based on user features provided in the above various embodiments.

[0185] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A page display method based on user characteristics, characterized in that: include: Acquire tag data of a user, where the tag data is used to represent features of the user; Calculating the similarity between the user and each pre-acquired user group according to the tag data; Each user group is used to represent a type of user; According to the similarity between the user and each user group, a plurality of target user groups are obtained, wherein the similarity between the plurality of target user groups and the user is greater than the similarity between the other user groups and the user, and the number of the plurality of target user groups is determined by the number of page modules that can be arranged on the recommendation page corresponding to the channel information accessed by the user; Acquire multiple page modules corresponding to the multiple target user groups; Displaying the multiple page modules on the recommendation page, wherein the recommendation page displays clustering attributes of the products in each page module; Each of the page modules includes information about at least one product, and the multiple page modules are arranged in descending order of similarity between the multiple target user groups and the user, with the page module corresponding to the target user group with the highest similarity being arranged at the front; The method further comprises: Configure page modules according to clustering attributes of multiple products to obtain multiple page modules; According to the characteristic information of each user group, a mapping relationship between the page module and the user group is established and stored, and the page module is an attribute page associated with the characteristics of the users in the user group.

2. The method according to claim 1, characterized in that The calculating the similarity between the user and each pre-acquired user group according to the tag data includes: For each user group, a cosine similarity calculation method is used to calculate the cosine similarity between the tag data and the feature information of the user group, and the similarity between the user and the user group includes the cosine similarity.

3. The method according to claim 1 or 2, characterized in that: The obtaining of multiple target user groups according to the similarity between the user and each user group includes: Sort the similarities between the user and each user group from large to small or from small to large, and obtain multiple target similarities with the largest similarity starting from the largest similarity; Acquire corresponding multiple target user groups according to the multiple target similarities.

4. The method according to claim 3, characterized in that The number of target user groups is determined based on the recommended pages.

5. The method according to claim 1 or 2, characterized in that: The step of obtaining the user's tag data includes: Upon receiving a page acquisition request carrying the identity information of the user, acquiring the historical consumption data of the user according to the identity information of the user; The tag data of the user is obtained according to the historical consumption data.

6. The method according to claim 1 or 2, characterized in that: The step of obtaining the user's tag data includes: Obtaining historical consumption data of a user according to the identity information of the user who subscribes to the page recommendation in the page recommendation subscription information; The tag data of the user is obtained according to the historical consumption data.

7. The method according to claim 1 or 2, characterized in that: The method further comprises: According to the channel information visited by the user, the recommendation page corresponding to the channel information is obtained.

8. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtaining a label feature vector for each user according to the annotated user data, wherein the label feature vector includes multiple labels corresponding to the user; Cluster users according to the feature vector of each user to obtain multiple user groups; For each user group, according to the feature vector of each user in the user group, a continuous bag of words (CBOW) model is used to calculate the feature information corresponding to the user group.

9. The method according to claim 1, characterized in that: The acquiring of a plurality of page modules corresponding to the plurality of target user groups comprises: According to the mapping relationship between the page modules and the user groups, a plurality of page modules corresponding to the plurality of target user groups are obtained.

10. A page display device based on user characteristics, characterized in that: include: An acquisition module, used to acquire tag data of a user, wherein the tag data is used to represent features of the user; A processing module, used for calculating the similarity between the user and each pre-acquired user group according to the tag data; each user group is used to represent a type of user; The processing module is further used to obtain multiple target user groups according to the similarity between the user and each user group, wherein the similarity between the multiple target user groups and the user is greater than the similarity between other user groups and the user, and the number of the multiple target user groups is determined by the number of page modules that can be arranged on the recommendation page corresponding to the channel information accessed by the user; The processing module is also used to obtain multiple page modules corresponding to the multiple target user groups; An interactive module, configured to display the plurality of page modules on the recommendation page, wherein the recommendation page displays clustering attributes of products in each page module; Each of the page modules includes information of a plurality of products, and the plurality of page modules are arranged in descending order of similarity between the plurality of target user groups and the user, with the page module corresponding to the target user group with the highest similarity being arranged at the front; the processing module is further used for: Configure page modules according to clustering attributes of multiple products to obtain multiple page modules; According to the characteristic information of each user group, a mapping relationship between the page module and the user group is established and stored, and the page module is an attribute page associated with the characteristics of the users in the user group.

11. An electronic device, characterized in that: include: processor, memory, and,interface; A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the page display method based on user characteristics as described in any one of claims 1 to 9 by executing the executable instructions.

12. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for displaying a page based on user features described in any one of claims 1 to 9 is implemented.

Images