A method, device, equipment and storage medium for pushing data within a target community

Abstract

The present invention provides a method, device, equipment and storage medium for pushing data in a target community. The method first obtains user data in the target community, and then processes the user data by calling the K-means clustering algorithm to group the users of the target community, wherein the user data includes user behavior data and user feature data; then, when it is determined that the user has entered an application, the user's device information is obtained, and a push service adapter is matched according to the device information; then, push data is selected based on the user's group, and the push data is sent to the user terminal through the push service adapter; finally, the user's feedback behavior on the push data is obtained, and the push data is dynamically adjusted based on the feedback behavior. The problem of poor recommendation effect due to large differences in push services of different equipment manufacturers is solved.

Description

Technical Field

[0001] The present invention relates to the field of data push, and in particular to a method, device, equipment and storage medium for pushing data within a target community. Background Art

[0002] With the popularization of the Internet and the widespread use of mobile devices, the data and information generated by users is increasing. In social networks, e-commerce and other information dissemination fields, users' demand for personalized push services is increasing. Existing push systems are insufficient in terms of the accuracy of message push and the security of content, and are often unable to effectively group and customize content push based on the personalized characteristics of users. In addition, due to the large differences in push services from different device manufacturers, such as Apple's APNs push service and Android device manufacturers' FCM push service, some devices may experience delays in receiving or even fail to receive push messages.

[0003] In view of this, this application is filed. Summary of the invention

[0004] The present invention discloses a data push method, device, equipment and storage medium within a target community, aiming to solve the problem of poor recommendation effect caused by large differences in push services of different equipment manufacturers.

[0005] The first embodiment of the present invention provides a method for pushing data in a target community, including:

[0006] Acquire user data in a target community, and call a K-means clustering algorithm to process the user data so as to group users in the target community, wherein the user data includes user behavior data and user feature data;

[0007] When it is determined that the user has entered the application, the user's device information is obtained, and a push service adapter is matched according to the device information;

[0008] Selecting push data based on the user group, and sending the push data to the user terminal through the push service adapter;

[0009] Acquire the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior.

[0010] Preferably, the step of obtaining user data in the target community and invoking a K-means clustering algorithm to process the user data to group the users of the target community is as follows:

[0011] Collect users in the target community offline through ETL to generate user feature data, and collect users in the target community in real time through Flume and Kafka to generate user behavior data;

[0012] Use the Pandas library to preprocess and feature process the user feature data and the user behavior data;

[0013] The users are clustered into different groups by using the K-means clustering algorithm on the user feature data and the user behavior data after preprocessing and feature processing.

[0014] Preferably, the acquiring of the user's device information when determining that the user has entered the application, and matching the push service adapter according to the device information, is specifically:

[0015] When it is determined that the user has entered the application, the device information of the user is identified, the best adaptation algorithm is called to operate on the device information to generate a list of available push service adapters, the list of available push service adapters is traversed, and the push service adapter with the highest score is selected;

[0016] The device information includes device brand, model, client version, communication operator, and operating system version;

[0017] The scores include a device compatibility score, a performance index score, and a historical performance score.

[0018] Preferably, after the push data is selected based on the user's group, the method further includes:

[0019] The push data is sent to the message dispatch center of the push service, and load balancing is performed through dynamic weight polling and consistent Hash algorithm, so that the server can intelligently allocate push tasks based on real-time load status and ensure that push tasks for the same user are processed by the same node.

[0020] The second embodiment of the present invention provides a data push device in a target community, including:

[0021] A user grouping unit, used to obtain user data in a target community, and call a K-means clustering algorithm to process the user data so as to group users in the target community, wherein the user data includes user behavior data and user feature data;

[0022] The adapter matching unit is used to obtain the user's device information when it is determined that the user enters the application, and match the push service adapter according to the device information;

[0023] A data push unit, configured to select push data based on the user group, and send the push data to the user terminal through the push service adapter;

[0024] A feedback unit is used to obtain the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior.

[0025] Preferably, the user grouping unit is specifically used for:

[0026] Collect users in the target community offline through ETL to generate user feature data, and collect users in the target community in real time through Flume and Kafka to generate user behavior data;

[0027] Use the Pandas library to preprocess and feature process the user feature data and the user behavior data;

[0028] The users are clustered into different groups by using the K-means clustering algorithm on the user feature data and the user behavior data after preprocessing and feature processing.

[0029] Preferably, the adapter matching unit is specifically used for:

[0030] When it is determined that the user has entered the application, the device information of the user is identified, the best adaptation algorithm is called to operate on the device information to generate a list of available push service adapters, the list of available push service adapters is traversed, and the push service adapter with the highest score is selected;

[0031] The device information includes device brand, model, client version, communication operator, and operating system version;

[0032] The scores include a device compatibility score, a performance index score, and a historical performance score.

[0033] Preferably, the data pushing unit is further used for:

[0034] The push data is sent to the message dispatch center of the push service, and load balancing is performed through dynamic weight polling and consistent Hash algorithm, so that the server can intelligently allocate push tasks based on real-time load status and ensure that push tasks for the same user are processed by the same node.

[0035] The third embodiment of the present invention provides a data push device within a target community, characterized in that it includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program can be executed by the processor to implement a data push method within a target community as described in any one of the above items.

[0036] A fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that a computer program is stored therein, and the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a method for pushing data within a target community as described in any one of the above items.

[0037] Based on the data push method, device, equipment and storage medium in a target community provided by the present invention, the user data in the target community is first obtained, and the K-means clustering algorithm is called to process the user data to group the users of the target community, wherein the user data includes user behavior data and user feature data; then, when it is determined that the user has entered the application, the user's device information is obtained, and the push service adapter is matched according to the device information; then, push data is selected based on the user's group, and the push data is sent to the user terminal through the push service adapter; finally, the user's feedback behavior on the push data is obtained, and the push data is dynamically adjusted based on the feedback behavior. The problem of poor recommendation effect due to large differences in push services of different equipment manufacturers is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flowchart of a method for pushing data within a target community provided by the first embodiment of the present invention;

[0039] Figure 2 It is a module diagram of a data push device within a target community provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0042] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0044] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0045] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0046] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0047] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0048] The present invention discloses a data push method, device, equipment and storage medium within a target community, aiming to solve the problem of poor recommendation effect caused by large differences in push services of different equipment manufacturers.

[0049] The first embodiment of the present invention provides a method for pushing data in a target community, which can be executed by a data pushing device (hereinafter referred to as a pushing device) in the target community, and in particular, by one or more processors in the pushing device to implement at least the following steps:

[0050] S101, obtaining user data in a target community, calling a K-means clustering algorithm to process the user data to group users in the target community, wherein the user data includes user behavior data and user feature data;

[0051] In this embodiment, the push device may be a terminal with data processing capabilities such as a server, workstation, desktop computer, laptop computer, etc. The push device may be installed with a corresponding operating system and application software, and the functions required by this embodiment are realized through the combination of the operating system and application software.

[0052] Specifically, in this embodiment, ETL (extraction, transformation and loading) technology can be used to collect offline data of users in the target community. The first step of the ETL process is to extract user data, which can come from multiple sources, such as personal information provided by users when registering, and historical interaction records of users on the platform. By transforming these data, such as cleaning and formatting, rich user feature data can be generated. These feature data will include basic information of users (such as age, gender, region), interests and hobbies, social behavior, and other variables that may affect user decisions.

[0053] Furthermore, in terms of real-time data collection, Apache Flume and Kafka are introduced in this embodiment, which can efficiently capture the behavior data of users in the target community. Flume is responsible for collecting user behavior data in real time from different data sources (such as application logs, event streams, etc.), and timely obtaining real-time interaction information of users in the application, such as click events, browsing time, sharing behavior, etc. Kafka is used as a high-throughput message queue to store and forward these behavior data.

[0054] After obtaining the user feature data and behavior data, use the Pandas library for preprocessing and feature processing. This process includes data cleaning, missing value processing, and feature engineering. For example, use mean filling or interpolation to handle missing data, and improve data comparability through standardization and normalization techniques. In addition, Pandas can also be used for feature selection and generation to extract features closely related to user behavior, such as user activity, interaction frequency, content preferences, etc.

[0055] In this embodiment, the K-means clustering algorithm can be applied to perform user clustering on the pre-processed and feature-processed data. Users are divided into multiple groups based on their features and behavior similarities. The clustering process first randomly selects K initial cluster centers, then calculates the distance between each user and the cluster center based on the user's feature data, and assigns the user to the cluster center with the closest distance. Through multiple iterations, the cluster center is continuously updated until convergence. It can not only effectively identify common features within the user group, but also discover potential segments for subsequent generation of user service solutions.

[0056] S102, when it is determined that the user enters the application, obtaining the user's device information, and matching the push service adapter according to the device information;

[0057] It should be noted that when it is determined that the user has entered the application, the user's device information will be quickly identified. Multiple key parameters are extracted from the user's device, such as device brand (such as Apple, Huawei, Samsung), model, client version, communication operator, and operating system version (such as different versions of Android and iOS). Then, the best adaptation algorithm is called to perform intelligent calculations on the collected device information. It can comprehensively consider multiple scoring indicators (device compatibility score, performance indicator score, historical performance score) to generate a list of available push service adapters. The device compatibility score will evaluate the degree of match between the selected adapter and the user's device to ensure that the message can be delivered without obstacles. For example, if the user is using an Android device of a specific brand, the system will give priority to push service adapters with strong compatibility with that brand. The performance indicator score measures the response speed and stability of each adapter in different network environments, and the historical performance score will be based on the adapter's past usage data, including message arrival rate and user interaction, to help screen out adapters with excellent performance.

[0058] After generating a list of available push service adapters, the push service adapter with the highest score is selected according to the above scoring criteria. It should be noted that in this embodiment, not only static scoring is relied upon, but also a dynamic feedback mechanism can be combined. Specifically, user feedback, such as the opening rate and click-through rate of push messages, can be monitored in real time to further optimize the adapter selection strategy. This feedback mechanism enables the push service to be continuously adjusted as user behavior changes.

[0059] S103, selecting push data based on the user group, and sending the push data to the user terminal through the push service adapter;

[0060] In this embodiment, users are divided into different groups through the previous K-means clustering algorithm, and each group represents a class of users with similar characteristics and behavior patterns. Therefore, the selection of push data will be based on the specific needs and preferences of each group. Specifically, the content that best matches the user characteristics of the group will be extracted from the push data pool. For example, users in a certain group may prefer shopping information, while users in another group may prefer to obtain social interaction messages. The selection process of push data is not only about matching static content, but also considers dynamic adjustments, such as adjusting the specific content of the push according to the user's real-time behavior or environment (such as the current time period or geographic location).

[0061] After selecting the push data, the data is sent to the user terminal through the intelligent push service adapter. The push service adapter has been selected in the previous step through the analysis of device information and the best adaptation algorithm, so it can ensure that the pushed content is delivered to the user terminal in the most appropriate way. Whether the user is using an Android device or an iOS device, the push adapter can optimize the push process based on the characteristics of the device (such as brand, model, operating system version, etc.).

[0062] After the push data is selected and the adapter is selected, it is pushed through the scheduling center, which can use dynamic weight polling and consistent hash algorithm for load balancing management. The dynamic weight polling mechanism is mainly to ensure that the push tasks can be intelligently allocated according to the real-time load of the server. For example, when some server nodes are in a high-load state, the system will automatically reduce the amount of tasks assigned to them and assign more tasks to nodes in a low-load state. This can effectively avoid server overload. The consistent hash algorithm further optimizes the allocation of push tasks. Specifically, the push tasks of the same user will always be processed by the same server node, which maintains the consistency of tasks between multiple push tasks, avoiding repeated push content and unnecessary resource consumption. At the same time, it can ensure the order of message push for those messages that rely on time order (such as chat notifications or transaction reminders).

[0063] S104: Obtain the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior.

[0064] In this embodiment, the push strategy can be optimized by generating recommended options to ensure that the push content received by each user not only meets their personal interests, but also dynamically adjusts the push strategy to improve the interaction effect. When selecting recommended options, the ε-greedy strategy in the MAB (multi-armed bandit) algorithm is used to dynamically balance the relationship between exploring new options and using known effective options.

[0065] First, the recommendation device generates multiple recommendation options for each user based on the type of content to be pushed (such as pictures, texts, videos, etc.). These options include different push content, different push times, and push methods. Use the ε-greedy strategy to select these recommendation options. Specifically, ε represents the exploration rate, that is, new recommendation options will be randomly selected with a certain probability, and 1-ε represents the utilization rate, that is, the best known recommendation option is selected based on the current push effect. By adjusting the ε value, the system can find the best balance between exploring new push strategies and utilizing the best known push strategies, ensuring that the push strategy can adapt to new scenarios and continuously optimize existing push solutions.

[0066] Furthermore, the recommendation device will dynamically adjust the weight value of each recommended option based on the historical performance of the option (such as the user's click rate, viewing time, subsequent operations, etc.). If a recommended option performs well among a certain type of user, its weight will gradually increase, and the system will select the recommended option more frequently; conversely, the recommended option with poor performance will reduce its chance of being selected. At the same time, the recommendation device will conduct more frequent exploration of new business scenarios and new recommended options that have not yet appeared in historical push notifications to quickly verify the effectiveness of these new options.

[0067] The recommendation device sets a basic weight for each user behavior, which is set according to the depth of interaction between the user and the push message. For example, the basic weight of viewing a push message is 0.8, the weight of clicking on a push message is 1.0, and the weight of entering the application is 1.2. In addition, the user's usage time in the application will also increase the weight, calculated in increments of 0.1 per minute. For more complex user behaviors (such as completing a task in multiple steps), the system will assign a weight to each step and accumulate these weights to accurately reflect the user's interaction depth. During the push process, user feedback data will be collected, such as the number of clicks, viewing time, usage time, and the number of times the application is re-entered after exiting. This feedback data will be used to adjust the weight of the push options, and this data will be integrated into the recommendation model to further optimize the push strategy. For example, the system can calculate the comprehensive weight of the recommended option through the following formula:

[0068] Weight calculation = viewing push weight + clicking push message weight + entering app weight + single usage time in minutes * 0.1 + (number of times you enter the app twice or more * weight after exiting and re-entering the app) + complex behavior weight. This calculation method ensures that the system can update the weights of recommended options based on the user's actual interactive behavior.

[0069] See also Figure 2 The second embodiment of the present invention provides a data push device in a target community, including:

[0070] A user grouping unit 201 is used to obtain user data in a target community and process the user data using a K-means clustering algorithm to group users in the target community, wherein the user data includes user behavior data and user feature data;

[0071] The adapter matching unit 202 is used to obtain the user's device information when it is determined that the user enters the application, and match the push service adapter according to the device information;

[0072] A data push unit 203, configured to select push data based on the user group, and send the push data to the user terminal through the push service adapter;

[0073] The feedback unit 204 is used to obtain the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior.

[0074] Preferably, the user grouping unit is specifically used for:

[0075] Collect users in the target community offline through ETL to generate user feature data, and collect users in the target community in real time through Flume and Kafka to generate user behavior data;

[0076] Use the Pandas library to preprocess and feature process the user feature data and the user behavior data;

[0077] The users are clustered into different groups by using the K-means clustering algorithm on the user feature data and the user behavior data after preprocessing and feature processing.

[0078] Preferably, the adapter matching unit is specifically used for:

[0079] When it is determined that the user has entered the application, the device information of the user is identified, the best adaptation algorithm is called to operate on the device information to generate a list of available push service adapters, the list of available push service adapters is traversed, and the push service adapter with the highest score is selected;

[0080] The device information includes device brand, model, client version, communication operator, and operating system version;

[0081] The scores include a device compatibility score, a performance index score, and a historical performance score.

[0082] Preferably, the data pushing unit is further used for:

[0083] The push data is sent to the message dispatch center of the push service, and load balancing is performed through dynamic weight polling and consistent Hash algorithm, so that the server can intelligently allocate push tasks based on real-time load status and ensure that push tasks for the same user are processed by the same node.

[0084] The third embodiment of the present invention provides a data push device within a target community, characterized in that it includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program can be executed by the processor to implement a data push method within a target community as described in any one of the above items.

[0085] A fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that a computer program is stored therein, and the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a method for pushing data within a target community as described in any one of the above items.

[0086] Based on the data push method, device, equipment and storage medium in a target community provided by the present invention, the user data in the target community is first obtained, and the K-means clustering algorithm is called to process the user data to group the users of the target community, wherein the user data includes user behavior data and user feature data; then, when it is determined that the user has entered the application, the user's device information is obtained, and the push service adapter is matched according to the device information; then, push data is selected based on the user's group, and the push data is sent to the user terminal through the push service adapter; finally, the user's feedback behavior on the push data is obtained, and the push data is dynamically adjusted based on the feedback behavior. The problem of poor recommendation effect due to large differences in push services of different equipment manufacturers is solved.

[0087] Exemplarily, the computer program described in the third and fourth embodiments of the present invention may be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the device for implementing data push in a target community. For example, the device described in the second embodiment of the present invention.

[0088] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the method for pushing data in a target community, and uses various interfaces and lines to connect the various parts of the method for pushing data in a target community.

[0089] The memory can be used to store the computer program and / or module, and the processor realizes various functions of a data push method in a target community by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0090] Wherein, if the implemented module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0091] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0092] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention 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 method for pushing data within a target community, characterized in that: include: Acquire user data in a target community, and call a K-means clustering algorithm to process the user data so as to group users in the target community, wherein the user data includes user behavior data and user feature data; When it is determined that the user has entered the application, the user's device information is obtained, and the push service adapter is matched according to the device information, specifically: when it is determined that the user has entered the application, the user's device information is identified, the best adaptation algorithm is called to operate on the device information to generate a list of available push service adapters, the list of available push service adapters is traversed, the push service adapter with the highest score is selected, and the user's feedback is obtained in real time to optimize the selection strategy of the push service adapter; wherein the device information includes the device brand, model, client version, communication operator, and operating system version; the score includes the device compatibility score, the performance index score, and the historical performance score, wherein the device compatibility score is used to evaluate the matching degree between the selected push service adapter and the user's device, the performance index score measures the response speed and stability of each adapter in different network environments, and the historical performance score is used to screen out excellent performance adapters based on the past usage data of the push service adapter, and the usage data includes message arrival rate and user interaction; Selecting push data based on the user group, and sending the push data to the user terminal through the push service adapter; Obtain the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior, wherein multiple recommendation options are generated for each user according to the type of content to be pushed, and the multiple recommendation options include different push content, different push times and push methods. Use an ε-greedy strategy to select the multiple recommendation options, and dynamically adjust the weight value of each recommendation option according to its historical performance.

2. A method for pushing data within a target community according to claim 1, characterized in that: The step of obtaining user data in the target community and invoking a K-means clustering algorithm to process the user data so as to group the users in the target community is as follows: Collect users in the target community offline through ETL to generate user feature data, and collect users in the target community in real time through Flume and Kafka to generate user behavior data; Use the Pandas library to preprocess and feature process the user feature data and the user behavior data; The users are clustered into different groups by using the K-means clustering algorithm on the user feature data and the user behavior data after preprocessing and feature processing.

3. The method for pushing data within a target community according to claim 1, characterized in that: After the push data is selected based on the user's group, the method further includes: The push data is sent to the message dispatch center of the push service, and load balancing is performed through dynamic weight polling and consistent Hash algorithm, so that the server can intelligently allocate push tasks based on real-time load status and ensure that push tasks for the same user are processed by the same node.

4. A data push device in a target community, characterized in that: include: A user grouping unit, used to obtain user data in a target community, and call a K-means clustering algorithm to process the user data so as to group users in the target community, wherein the user data includes user behavior data and user feature data; An adapter matching unit, used to obtain the user's device information when it is determined that the user has entered the application, and match the push service adapter according to the device information, specifically used to: identify the user's device information when it is determined that the user has entered the application, call the best adaptation algorithm to operate on the device information to generate a list of available push service adapters, traverse the list of available push service adapters, select the push service adapter with the highest score, and obtain user feedback in real time to optimize the selection strategy of the push service adapter; wherein the device information includes device brand, model, client version, communication operator, and operating system version; the score includes device compatibility score, performance index score, and historical performance score, wherein the device compatibility score is used to evaluate the matching degree between the selected push service adapter and the user's device, the performance index score measures the response speed and stability of each adapter in different network environments, and the historical performance score is used to screen out excellent performance adapters based on the past usage data of the push service adapter, and the usage data includes message arrival rate and user interaction; A data push unit, configured to select push data based on the user group, and send the push data to the user terminal through the push service adapter; A feedback unit is used to obtain the user's feedback behavior on the pushed data, and dynamically adjust the pushed data based on the feedback behavior, wherein multiple recommendation options are generated for each user according to the type of content to be pushed, and the multiple recommendation options include different push content, different push time and push method, and an ε-greedy strategy is used to select the multiple recommendation options, and the weight value of each recommendation option is dynamically adjusted according to the historical performance of the option.

5. The data push device in a target community according to claim 4, characterized in that: The user grouping unit is specifically used for: Collect users in the target community offline through ETL to generate user feature data, and collect users in the target community in real time through Flume and Kafka to generate user behavior data; Use the Pandas library to preprocess and feature process the user feature data and the user behavior data; The users are clustered into different groups by using the K-means clustering algorithm on the user feature data and the user behavior data after preprocessing and feature processing.

6. The data push device in a target community according to claim 4, characterized in that: The data push unit is also used for: The push data is sent to the message dispatch center of the push service, and load balancing is performed through dynamic weight polling and consistent Hash algorithm, so that the server can intelligently allocate push tasks based on real-time load status and ensure that push tasks for the same user are processed by the same node.

7. A data push device in a target community, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a method for pushing data within a target community as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a method for pushing data within a target community as described in any one of claims 1 to 3.

Images