Intelligent data pushing method based on user behavior analysis
Through the method based on user behavior analysis, the matching degree calculation of data labels and user labels is used to calculate the data labels and user labels, which solves the problem of inaccurate data push in the prior art and improves the efficiency of information dissemination and user experience.
Patent Information
- Application Number
- CN202410659087.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-27
- Publication Date
- 2025-01-07
AI Technical Summary
The existing heterogeneous graph-based data push method cannot effectively push data to designated groups as expected, resulting in low information dissemination efficiency and users may receive meaningless information, causing users to dislike and messages that are really needed to be blocked.
Using a method based on user behavior analysis, the data analysis module is used to analyze the push data, matched data labels and user labels are obtained, feature extraction and matching degree calculation are used using neural network models, and whether to push data is determined based on the matching degree threshold.
It improves the accuracy of data push, reduces the spam information received by the user, and ensures the timely communication of important information.
Smart Images
Figure CN119271872A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent data push, and in particular to a method for intelligent data push based on user behavior analysis. Background Art
[0002] With the rapid development of the Internet industry, massive amounts of data are pushed to users, but the data pushed to users may not be of interest to the users. Being able to push personalized data to different users is an effective way to improve user stickiness; but currently, the personalized data push to different users generally adopts a push method based on heterogeneous graphs. This method constructs the data identifiers of the data to be pushed and the user tags of the users into a heterogeneous graph, and uses graph embedding algorithms (such as metapath2vec, node2vec, etc.) to learn the heterogeneous graph, and then determines whether to push the data to the user. However, since this method only processes the data identifiers of the data to be pushed and pushes it, it cannot push the data to the specified group as expected, resulting in low information dissemination efficiency. In addition, meaningless information may cause users to be disgusted, thereby blocking the push messages, which may cause users to be unable to receive the push messages they really need. Summary of the invention
[0003] The purpose of this application is to provide a data intelligent push method based on user behavior analysis, which can improve the accuracy of data push and reduce junk information received by the user end.
[0004] To achieve the above objectives:
[0005] In a first aspect, an embodiment of the present application provides a method for intelligent data push based on user behavior analysis, the method comprising:
[0006] In response to receiving the data to be pushed, triggering the data analysis module to analyze the data to be pushed and obtain a data tag matching the data to be pushed;
[0007] Acquire a user tag that matches the data tag, and acquire user data under the user tag;
[0008] The matching degree between the data to be pushed and the user under the user tag is calculated based on the user data, and when the matching degree is greater than a preset matching degree threshold, the data to be pushed is pushed to the user under the user tag.
[0009] In one embodiment, obtaining a user tag matching the data tag and obtaining user data under the user tag includes:
[0010] Extracting features from the data tags through a neural network model, and using a back propagation algorithm to obtain user tags that match the features of the data tags;
[0011] Based on the user tag, user data satisfying the user tag is acquired.
[0012] In one embodiment, before extracting features from the data tags through a neural network model, the method includes: training the neural network model through training samples so that the neural network model has the ability to match the data tags with the user tags.
[0013] In one embodiment, the training of the neural network model by training samples includes:
[0014] Acquire training samples; perform text encoding processing on the training key description information in the training samples through a neural network model to obtain a feature vector of the training data described by the training key description information; perform conversion processing on the training label vector corresponding to the training key description information to obtain a label vector corresponding to the user label; process the feature vector of the training data described by the training key description information and the label vector corresponding to the training label through a loss function to obtain a loss value corresponding to the training sample; train the preset deep learning model based on the loss value corresponding to the training sample to obtain the neural network model.
[0015] In one embodiment, after obtaining the user tag matching the data tag and obtaining the user data under the user tag, the method further includes:
[0016] The historical access data in the user data is obtained, and the users are classified according to the historical access data to obtain target users.
[0017] In one implementation, the calculating, based on the user data, the matching degree between the data to be pushed and the user under the user tag includes:
[0018] Access data of the target user to data of the same type as the data to be pushed is obtained, and the interest value of the target user to the data to be pushed is calculated based on the access data; and the matching degree between the data to be pushed and the target user is determined according to the interest value.
[0019] In one embodiment, the data intelligent push method based on user behavior analysis further includes: obtaining a data operation log of a target user, and determining historical operation behavior information of the target user on various tags based on the data operation log;
[0020] A user tag is set for the target user according to the operation behavior information; the user tag includes data tag information of various types of test data in the operation behavior information.
[0021] In a second aspect, an embodiment of the present application provides a data intelligent push device based on user behavior analysis, the device comprising:
[0022] A data analysis module, configured to analyze the data to be pushed in response to receiving the data to be pushed, and obtain a data tag matching the data to be pushed;
[0023] A tag matching module, used to obtain a user tag that matches the data tag, and obtain user data under the user tag;
[0024] The data push module is used to calculate the matching degree between the data to be pushed and the user under the user tag based on the user data, and push the data to be pushed to the user under the user tag when the matching degree is greater than a preset matching degree threshold.
[0025] In a third aspect, the present application provides an electronic device, including: a processor;
[0026] a memory for storing instructions executable by the processor;
[0027] The processor is configured to execute the instructions to implement the data intelligent push method based on user behavior analysis as described in any one of the first aspects.
[0028] In a fourth aspect, the present application provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor, the data intelligent push method based on user behavior analysis as described in any one of the first aspects is implemented.
[0029] The present application provides a method for intelligent data push based on user behavior analysis, which includes: in response to receiving data to be pushed, triggering a data analysis module to analyze the data to be pushed, and obtaining a data tag matching the data to be pushed; obtaining a user tag matching the data tag, and obtaining user data under the user tag; calculating the matching degree between the data to be pushed and the user under the user tag based on the user data, and when the matching degree is greater than a preset matching degree threshold, pushing the data to be pushed to the user under the user tag. In this way, the data to be pushed can be pushed to a specified group of people according to the data type, improving the accuracy of data push, and reducing the spam received by the user end. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1A schematic diagram of a process for intelligently pushing data based on user behavior analysis provided by an embodiment of the present invention;
[0031] Figure 2 A schematic diagram of the structure of a data intelligent push device based on user behavior analysis provided by an embodiment of the present invention;
[0032] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0033] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0034] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.
[0035] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of..." or "when..." or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising", "including" indicate that there are described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, “A, B, or C” or “A, B and / or C” means “any of the following: A; B; C; A and B; A and C; B and C; A, B, and C.” An exception to this definition will occur only when a combination of elements, functions, steps, or operations are inherently mutually exclusive in some manner.
[0036] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
[0037] It should be noted that in this article, step codes such as S1 and S2 are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the order. When implementing the step, those skilled in the art may execute S2 first and then S1, etc., but these should all be within the scope of protection of this application.
[0038] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0039] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.
[0040] See also Figure 1 , is a flow chart of a method for intelligent data push based on user behavior analysis provided in an embodiment of the present application. The method for intelligent data push based on user behavior analysis can be executed by an electronic device provided in an embodiment of the present application. The electronic device can be implemented in software and / or hardware, such as a server, a computer, or other electronic device. In this embodiment, the execution subject of the method for intelligent data push based on user behavior analysis is a server as an example. The method for intelligent data push based on user behavior analysis provided in this embodiment includes:
[0041] Step S1, in response to receiving the data to be pushed, triggering the data analysis module to analyze the data to be pushed and obtain a data tag matching the data to be pushed.
[0042] Here, a large number of data tags are stored in the database on the server side. When the server side receives data that needs to be pushed to the user side, it automatically triggers the data intelligent push device based on user behavior analysis to analyze the data to be pushed and match the data to be pushed with the data tags. By matching the data to be pushed with the data tags, the data to be pushed can be classified by tags, thereby narrowing the user range of the push and reducing the amount of calculation in the subsequent data intelligent push process based on user behavior analysis.
[0043] Step S2, obtaining a user tag that matches the data tag, and obtaining user data under the user tag.
[0044] It is understandable that data tags and user tags may include multiple tags such as digital, finance, history, etc. By pulling user data under the user tag and then analyzing the user data again, the push scope can be narrowed down, making it easier to accurately find users who match the data to be pushed.
[0045] In one embodiment, the obtaining of a user tag that matches the data tag and obtaining user data under the user tag includes: extracting features of the data tag through a neural network model and obtaining a user tag that matches the features of the data tag using a back propagation algorithm; and obtaining user data that satisfies the user tag based on the user tag.
[0046] Here, the back propagation algorithm can calculate the error of the output layer, that is, the difference between the data label and the user label, and then return the calculated difference to each layer of the neural network. Through the propagation of the difference, the algorithm can calculate the error contribution of each neuron. Based on this, the neural network model will update each connection weight to reduce the error contribution of each neuron. In this way, the neural network is optimized by adjusting the weights so that the output of the network (user label) is as close to the expected output (the characteristics of the data label) as possible. In this way, when the neural network runs on new user data, it can accurately identify user labels that match the characteristics learned during training, thereby obtaining user data that meets these user labels to improve the accuracy of user-data matching.
[0047] In one embodiment, the neural network model is trained through training samples so that the neural network model has the ability to match the data label with the user label.
[0048] In one embodiment, the training of the neural network model through training samples includes: obtaining training samples; performing text encoding processing on the training key description information in the training samples through the neural network model to obtain the feature vector of the training data described by the training key description information; converting the training label vector corresponding to the training key description information to obtain the label vector corresponding to the user label; processing the feature vector of the training data described by the training key description information and the label vector corresponding to the training label through a loss function to obtain the loss value corresponding to the training sample; training the preset deep learning model based on the loss value corresponding to the training sample to obtain the neural network model.
[0049] Here, the training samples contain text data and corresponding labels. These samples are used to train the neural network model so that it can learn how to extract features from text and map these features to the correct labels. The neural network model then encodes the text data in the training samples. This involves vectorization of the text, that is, converting the text into a numerical form that can be processed by the neural network. This process includes word embeddings, where each word or phrase is converted into a fixed-size vector that captures the semantic information of the vocabulary. The labels of the training samples also need to be converted into numerical form, usually vectors, so that the neural network can process them. This process involves one-hot encoding, where each label is converted into a vector in which only the elements corresponding to the label are 1 and the rest are 0. The loss function is used to evaluate the difference between the prediction results of the neural network and the actual labels. During the training process, the neural network will try to minimize this loss value. The loss function can take many forms, such as cross-entropy loss. Based on the calculated loss value, the neural network model is trained through an optimization algorithm (such as stochastic gradient descent). During this process, the parameters (weights and biases) of the model are updated to reduce the loss value. The training process usually includes multiple iterations, each of which performs a forward propagation and a backward propagation on the entire training set to gradually improve the performance of the model. After enough training iterations, the neural network model can extract key features from the text and map these features to the correct user tags. In this way, the accuracy of the neural network model in identifying tags can be improved, thereby reducing the amount of calculation in the subsequent process and accurately pushing data to the target population.
[0050] In one embodiment, after obtaining the user tag matching the data tag and obtaining the user data under the user tag, the method further includes: obtaining historical access data in the user data, classifying the user according to the historical access data, and obtaining the target user.
[0051] Here, the user's historical access data includes the web pages visited by the user, the applications used, the transactions made, the advertisements clicked, etc. By extracting useful features from the historical access data, such as user access frequency, access time, length of stay, and user behavior patterns, the users are classified according to these features, and then the target users are identified based on the classification results. By classifying users through historical access data, the target users can be identified more accurately and effectively, and the accuracy of data push can be improved.
[0052] In one embodiment, the calculation of the matching degree between the data to be pushed and the user under the user tag based on the user data includes: obtaining the access data of the target user to the same type of data to be pushed, and calculating the interest value of the target user in the data to be pushed based on the access data; and determining the matching degree between the data to be pushed and the target user according to the interest value.
[0053] It can be understood that after the target user is determined through historical access data, the historical access data of the target user is screened again to extract data of the same type as the data to be pushed, and these data are counted. The target user's interest value in the pushed data is determined based on the proportion of data of the same type as the data to be pushed in the target user's historical access data. This can further narrow the target audience of the data to be pushed and achieve the purpose of accurately pushing data.
[0054] In one embodiment, the data intelligent push method based on user behavior analysis also includes: obtaining the data operation log of the target user, and determining the historical operation behavior information of the target user for various tags based on the data operation log; setting a user tag for the target user according to the operation behavior information; the user tag contains data tag information of various types of test data in the operation behavior information.
[0055] Here, by determining the user tag through the user's data operation log, the users can be accurately divided into different groups, and then it is determined whether to push the data to be pushed for different groups, thereby realizing targeted data push.
[0056] The method for intelligent data push based on user behavior analysis provided by the embodiment of the present invention matches the data to be pushed with the user's label, and through multi-level identification of users, gradually narrows the target user range. The data to be pushed can be pushed to a designated group of people according to the data type, thereby improving the accuracy of data push and reducing the spam information received by the user end.
[0057] Based on the same inventive concept as the above embodiments, the embodiments of the present invention provide a data intelligent push device based on user behavior analysis, such as Figure 2 As shown, the device comprises:
[0058] A data analysis module, configured to analyze the data to be pushed in response to receiving the data to be pushed, and obtain a data tag matching the data to be pushed;
[0059] A tag matching module, used to obtain a user tag that matches the data tag, and obtain user data under the user tag;
[0060] The data push module is used to calculate the matching degree between the data to be pushed and the user under the user tag based on the user data, and push the data to be pushed to the user under the user tag when the matching degree is greater than a preset matching degree threshold.
[0061] Based on the same inventive concept as the above embodiments, an embodiment of the present invention provides an electronic device, such as Figure 3 As shown, the electronic device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 3 The processor 310 shown in the figure is not used to indicate that the number of the processor 310 is one, but is only used to indicate the positional relationship of the processor 310 relative to other devices. In actual applications, the number of the processor 310 may be one or more; similarly, Figure 3 The memory 311 shown in the figure has the same meaning, that is, it is only used to refer to the position relationship of the memory 311 relative to other devices. In practical applications, the number of memories 311 can be one or more. When the processor 310 runs the computer program, the above-mentioned data intelligent push method based on user behavior analysis is implemented.
[0062] The electronic device may also include: at least one network interface 312. The various components in the electronic device are coupled together via a bus system 313. It is understood that the bus system 313 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the bus system 313 is not used in the following examples. Figure 3 Various buses are labeled as bus system 313.
[0063] The memory 311 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. The nonvolatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), direct memory bus random access memory (DRRAM). The memory 311 described in the embodiments of the present invention is intended to include but is not limited to these and any other suitable types of memory.
[0064] The memory 311 in the embodiment of the present invention is used to store various types of data to support the operation of the electronic device. Examples of such data include: any computer program used to operate on the electronic device.
[0065] Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a computer-readable storage medium, in which a computer program is stored. The computer-readable storage medium may be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); or various devices including one or any combination of the above-mentioned memories, such as a server. When the instructions stored therein are executed by a processor, the data intelligent push method based on user behavior analysis described in the above-mentioned embodiment is implemented.
[0066] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described 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.
[0067] In this document, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than those listed and may also include additional elements not expressly listed.
[0068] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A data intelligent push method based on user behavior analysis, characterized in that: include: In response to receiving the data to be pushed, triggering the data analysis module to analyze the data to be pushed and obtain a data tag matching the data to be pushed; Acquire a user tag that matches the data tag, and acquire user data under the user tag; The matching degree between the data to be pushed and the user under the user tag is calculated based on the user data, and when the matching degree is greater than a preset matching degree threshold, the data to be pushed is pushed to the user under the user tag.
2. The method according to claim 1, characterized in that The obtaining of the user tag matching the data tag and obtaining the user data under the user tag includes: Extracting features from the data tags through a neural network model, and using a back propagation algorithm to obtain user tags that match the features of the data tags; Based on the user tag, user data satisfying the user tag is acquired.
3. The method according to claim 2, characterized in that Before extracting features from the data labels using a neural network model, the method includes: The neural network model is trained through training samples so that the neural network model has the ability to match the data label with the user label.
4. The method according to claim 3, characterized in that: The step of training the neural network model by using training samples includes: Get training samples; Performing text encoding processing on the training key description information in the training sample through a neural network model to obtain a feature vector of the training data described by the training key description information; Converting the training label vector corresponding to the training key description information to obtain a label vector corresponding to the user label; Processing the feature vector of the training data described by the training key description information and the label vector corresponding to the training label by a loss function to obtain a loss value corresponding to the training sample; The preset deep learning model is trained based on the loss value corresponding to the training sample to obtain the neural network model.
5. The method according to claim 1, characterized in that After obtaining the user tag matching the data tag and obtaining the user data under the user tag, the method further includes: The historical access data in the user data is obtained, and the users are classified according to the historical access data to obtain target users.
6. The method according to claim 5, characterized in that The calculating, based on the user data, the matching degree between the data to be pushed and the user under the user tag includes: Acquire access data of the target user to data of the same type as the data to be pushed, and calculate the interest value of the target user to the data to be pushed based on the access data; The matching degree between the data to be pushed and the target user is determined according to the interest value.
7. The method according to claim 5, characterized in that The method further comprises: Obtain the data operation log of the target user, and determine the historical operation behavior information of the target user on various tags based on the data operation log; A user tag is set for the target user according to the operation behavior information; the user tag includes data tag information of various types of test data in the operation behavior information.
8. A data intelligent push device based on user behavior analysis, characterized in that: include: A data analysis module, configured to analyze the data to be pushed in response to receiving the data to be pushed, and obtain a data tag matching the data to be pushed; A tag matching module, used to obtain a user tag that matches the data tag, and obtain user data under the user tag; The data push module is used to calculate the matching degree between the data to be pushed and the user under the user tag based on the user data, and push the data to be pushed to the user under the user tag when the matching degree is greater than a preset matching degree threshold.
9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the data intelligent push method based on user behavior analysis as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor, the data intelligent push method based on user behavior analysis as described in any one of claims 1 to 7 is implemented.