Intelligent pushing system and method based on user habits

Abstract

The invention relates to the technical field of intelligent pushing, and discloses an intelligent pushing system and method based on user habitues.The method comprises the steps of obtaining target information of a target user, forming analysis data, forming pushing data, extracting corresponding content in a database and pushing the content to the target user. According to the browsing selection of the user, the preference of the user is analyzed and judged, and the specific browsing preference of the user is determined according to the browsing condition of the user, so that more really interested contents are accurately pushed to the user, the condition of inaccurate pushing is reduced, the browsing experience of the user is improved, the user can browse more favorite contents, and the user experience is improved. More users are attracted, and more users can stay for a long time.

Description

Technical Field

[0001] The present invention relates to the field of intelligent push technology, and in particular to an intelligent push system and method based on user habits. Background Art

[0002] The intelligent push system based on user habits is a system that uses user behavior data and habits to provide personalized content or services. Through accurate push, it improves user experience and system effectiveness. The system includes: data collection module, data processing and analysis module, intelligent push algorithm, push execution module, feedback and optimization module, etc. It can significantly improve user satisfaction and participation, while increasing the system's operational efficiency and benefits;

[0003] Record user behavior data, collect user preference information, process raw data, remove noise, and organize it into a usable format; establish user portraits based on user behavior data and preference information; use machine learning and data mining techniques to identify user behavior patterns and predict users' future needs; use algorithms to generate personalized recommendations based on user portraits and behavior patterns; adjust push strategies based on users' real-time feedback and new behavior data to ensure the relevance and accuracy of recommendations; generate personalized push content based on recommendation algorithms; select appropriate push channels; control the frequency of push to avoid disturbing users too much; collect user feedback on push content; analyze push effects; optimize recommendation algorithms and push strategies to improve the overall performance of the system;

[0004] Integrate data collection tools to ensure that users' behavior data and preference information can be collected in real time or regularly. Use data mining technology to analyze users' historical behavior, create detailed user portraits, select appropriate recommendation algorithms, and develop and optimize algorithms. Integrate the push system into the existing platform, conduct system testing and performance evaluation to ensure the accuracy and effectiveness of push content. Set up user feedback channels, regularly collect and analyze user feedback, and continuously optimize push strategies and system performance.

[0005] Existing intelligent push systems and methods based on user habits cannot analyze and judge user preferences based on user browsing choices, and cannot determine user specific browsing preferences based on user browsing conditions, which can easily lead to inaccurate push, reduce user browsing experience, and prevent users from browsing more content that they are truly interested in. There are certain limitations to its practicality. Summary of the invention

[0006] The present invention provides an intelligent push system and method based on user habits, which are used to promote the solution of the problems mentioned in the background technology.

[0007] The present invention provides the following technical solution: an intelligent push method based on user habits, comprising:

[0008] Obtain target information of target users;

[0009] The target information includes the target user's browsing content and operations on the browsing content;

[0010] The operations include browsing operations and deleting operations;

[0011] Based on the target information, through information analysis strategies, analytical data is formed to determine the user's preferences for browsing content, so as to facilitate more accurate push of browsing content that the target user is interested in;

[0012] According to the analysis data, push data is formed through target adjustment strategy or user preference determination strategy to determine the preferences of target users, and the reasons why users perform deletion operations on browsing contents are further determined, so as to facilitate more accurate push of browsing contents of interest to target users;

[0013] According to the push data, the corresponding content in the database is extracted and pushed to the target user, so as to more accurately push the browsing content that the target user is interested in.

[0014] As a user habit-based intelligent push method described in the present invention, the information analysis strategy is specifically:

[0015] Get the target user's current browsing content;

[0016] Get the target user's operation on the currently browsed content;

[0017] If the target user's operation on the current browsing content is a browsing operation, then the browsing time of the target user on the current browsing content is obtained and determined as the user browsing time;

[0018] Get the total duration of the current browsing content and set it as the content duration;

[0019] Calculate the user browsing ratio, user browsing ratio = user browsing time ÷ content time;

[0020] Set the duration ratio threshold;

[0021] If the user browsing ratio is ≥ the duration ratio threshold, the target adjustment strategy is executed;

[0022] If the user browsing ratio is less than the duration ratio threshold, no judgment is made;

[0023] If the target user's operation on the currently browsed content is a delete operation, the user preference determination strategy is executed.

[0024] As a user habit-based intelligent push method described in the present invention, the target adjustment strategy is specifically:

[0025] Get the current browsing content;

[0026] Obtain all browsing contents in the database and set them as target browsing contents;

[0027] If there is a target browsing content identical to the current browsing content in the database, the target browsing content is identified as the target push content;

[0028] If the target browsing content identical to the current browsing content does not exist in the database, then the content keyword of the current browsing content is obtained;

[0029] Extract all target browsing contents containing content keywords from the database and identify them as target push contents;

[0030] Integrate all target push content into a push collection;

[0031] Extract any target push content in the push collection as the next current browsing content of the target user.

[0032] As a user habit-based intelligent push method described in the present invention, the user preference determination strategy is specifically:

[0033] Obtaining target data of the currently browsed content, wherein the target data includes title, author and content body;

[0034] Get all browsed content in the database;

[0035] Get the title of the currently browsed content and set it as the target title;

[0036] The browsing contents in the database with the same title as the target title are defined as the target judgment contents;

[0037] If there is target determination content in the database, all target determination content is extracted to form a target determination data set;

[0038] If the target judgment content does not exist in the database, the keywords of the target title are extracted and determined as the judgment keywords;

[0039] Extract all browsing contents containing judgment keywords in the database, identify them as target judgment contents, and form a target judgment data set;

[0040] Set the number of push notifications;

[0041] All target judgment contents in the target judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes;

[0042] Obtain the target user's operation on each target determination content and define it as the target determination operation;

[0043] Extracting the number of target determination contents whose target determination operation is a deletion operation, and setting it as a first determination number;

[0044] If the first judgment number ÷ the judgment push number ≥ 60%, the target title and browsing content related to the judgment keyword will no longer be pushed to the target user;

[0045] Mark all target judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database;

[0046] Extract any browsing content in the push database as the next current browsing content of the target user;

[0047] If the first determination number ÷ the determined push number < 60%, the user determination strategy is executed.

[0048] As a user habit-based intelligent push method described in the present invention, the user determination strategy is specifically:

[0049] Get all browsed content in the database;

[0050] Get the author ID of the currently browsed content and set it as the target ID;

[0051] The browsing content associated with the target ID in the database is defined as the associated determination content;

[0052] If there is association determination content in the database, all the association determination content is extracted to form an association determination data set;

[0053] If the associated determination content does not exist in the database, executing the first content determination strategy;

[0054] Get the number of push notifications;

[0055] All the related judgment contents in the related judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes;

[0056] Obtaining the target user's operation on each relevance determination content, and defining it as a relevance determination operation;

[0057] Extracting the number of associated determination contents whose associated determination operation is a deletion operation, and setting it as a second determination number;

[0058] If the second determined number ÷ the determined push number ≥ 60%, the browsing content associated with the target ID will no longer be pushed to the target user;

[0059] Mark all relevant judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database;

[0060] Extract any browsing content in the push database as the next current browsing content of the target user;

[0061] If the second determined number ÷ the determined push number < 60%, the first content determination strategy is executed.

[0062] As a user habit-based intelligent push method described in the present invention, wherein: the first content determination strategy is specifically:

[0063] Get all browsed content in the database;

[0064] Acquire the content body of the currently browsed content, wherein the content body includes a first body;

[0065] Obtain the first main body of the currently browsed content and set it as the first target content;

[0066] The browsing content having the first target content in the database is determined as the first determination content;

[0067] If the first determination content exists in the database, extract all the first determination content to form a first determination data set;

[0068] If the first content determination strategy does not exist in the database, the second content determination strategy is executed;

[0069] Get the number of push notifications;

[0070] Push all the first judgment contents in the first judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0071] Obtaining an operation of the target user on each first determination content as a first operation;

[0072] Extracting the number of first determination contents whose first operation is a deletion operation and setting it as a third determination number;

[0073] If the third determined number ÷ the determined push number ≥ 60%, the browsing content containing the first target content will no longer be pushed to the target user;

[0074] Marking all first determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0075] Extract any browsing content from the push database as the next current browsing content of the target user;

[0076] If the third determination number ÷ the determination push number < 60%, the second content determination strategy is executed.

[0077] As a user habit-based intelligent push method described in the present invention, the second content determination strategy is specifically:

[0078] Get all browsed content in the database;

[0079] Acquire the content body of the currently browsed content, wherein the content body includes a second body;

[0080] Obtain the second main body of the currently browsed content and set it as the second target content;

[0081] The browsing content having the second target content in the database is determined as the second determination content;

[0082] If the second determination content exists in the database, extract all the second determination content to form a second determination data set;

[0083] If the second determination content does not exist in the database, the third content determination strategy is executed;

[0084] Get the number of push notifications;

[0085] Push all the second judgment contents in the second judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0086] Obtaining an operation of the target user on each second determination content as a second operation;

[0087] Extracting the number of second determination contents in which the second operation is a deletion operation, and setting the number as a fourth determination number;

[0088] If the fourth determination number ÷ the determination push number ≥ 60%, the browsing content containing the second target content will no longer be pushed to the target user;

[0089] Marking all second determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0090] Extract any browsing content in the push database as the next current browsing content of the target user;

[0091] If the fourth determination number ÷ the determined push number < 60%, the third content determination strategy is executed.

[0092] As a user habit-based intelligent push method described in the present invention, the third content determination strategy is specifically:

[0093] Get all browsed content in the database;

[0094] Acquire the content body of the currently browsed content, wherein the content body includes a third body;

[0095] Obtain the third subject of the currently browsed content and define it as the third target content;

[0096] The browsing content having the third target content in the database is defined as the third determination content;

[0097] If the third determination content exists in the database, extract all the third determination content to form a third determination data set;

[0098] If the third determination content does not exist in the database, it is determined that the user has made an accidental touch operation, and any browsing content in the database is extracted as the next current browsing content of the target user;

[0099] Get the number of push notifications;

[0100] Push all the third judgment contents in the third judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0101] Obtaining an operation of the target user on each third determination content as a third operation;

[0102] Extracting the number of third determination contents whose third operation is a deletion operation, and setting it as a fifth determination number;

[0103] If the fifth determination number ÷ the determination push number ≥ 60%, the browsing content containing the third target content will no longer be pushed to the target user;

[0104] Marking all third-determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0105] Extract any browsing content in the push database as the next current browsing content of the target user;

[0106] If the fifth determination number ÷ the determination push number < 60%, it is determined that the user has made an accidental touch operation, and a browsing content in the database is randomly extracted as the next current browsing content of the target user.

[0107] The present invention also discloses a system for executing an intelligent push method based on user habits, which includes:

[0108] Acquisition module: obtain target information of target users;

[0109] Analysis module: Based on the target information, through information analysis strategy, analysis data is formed to determine the user's preferences for browsing content, so as to facilitate more accurate push of browsing content that the target user is interested in;

[0110] Judgment module: Based on the analysis data, through the target adjustment strategy or user preference judgment strategy, push data is formed to determine the preferences of the target user, and for the browsing content with deletion operation, the reason why the user performs the operation is further determined, so as to facilitate more accurate push of the browsing content that the target user is interested in;

[0111] Push module: extracts the corresponding content from the database based on the push data, and pushes the content to the target user, so as to more accurately push the browsing content that the target user is interested in.

[0112] The present invention has the following beneficial effects:

[0113] 1. The intelligent push system and method based on user habits determines the user's preference for the currently browsed content by obtaining the user's operation on the currently browsed content. If the user browses for a long time, it means that the user likes this type of video more. Then, the relevant videos will continue to be pushed to the user, thereby accurately pushing more content that the user is really interested in, reducing the situation of inaccurate push, improving the user's browsing experience, enabling the user to browse more content they like, attracting more users, and enabling more users to retain for a long time.

[0114] 2. The intelligent push system and method based on user habits determines the user's preference for the currently browsed content by obtaining the user's operation on the currently browsed content. If the user chooses to delete the current video, it means that the user does not like this type of video, then the push of related videos to the user will be stopped, thereby accurately pushing more content that the user is truly interested in, reducing the situation of inaccurate push, improving the user's browsing experience, enabling the user to browse more content they like, attracting more users, and enabling more users to retain for a long time.

[0115] 3. The intelligent push system and method based on user habits obtains relevant data of the currently browsed content, extracts all contents related to the currently browsed content in the database, pushes them to the user in sequence, obtains the user's operation, determines the reason why the user deleted the currently browsed content, and if the user is mostly interested in videos related to the currently browsed content, it means that the user has made an accidental touch operation. There is no need to adjust the pushed content, reduce the situation of inaccurate push, improve the user's browsing experience, enable users to browse more content they like, attract more users, and enable more users to retain for a long time. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] Figure 1This is a flow chart of the intelligent push method based on user habits of the present invention;

[0117] Figure 2 A system block diagram of the present invention for executing the intelligent push method based on user habits. DETAILED DESCRIPTION

[0118] 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.

[0119] Embodiment 1, a method for intelligent push based on user habits, see Figure 1 ,include:

[0120] Obtain target information of target users;

[0121] The target information includes the target user's browsing content and operations on the browsing content;

[0122] The operation includes a browsing operation and a deleting operation. The browsing operation is when a user browses a web page or video content, including browsing a comment area and browsing the homepage of the creator of the content, etc. The deleting operation is when a user chooses not to like the web page or video content;

[0123] Based on the target information, through information analysis strategies, analytical data is formed to determine the user's preferences for browsing content, so as to facilitate more accurate push of browsing content that the target user is interested in;

[0124] According to the analysis data, push data is formed through target adjustment strategy or user preference determination strategy to determine the preferences of target users, and the reasons why users perform deletion operations on browsing contents are further determined, so as to facilitate more accurate push of browsing contents of interest to target users;

[0125] According to the push data, the corresponding content in the database is extracted and pushed to the target user, so as to more accurately push the browsing content that the target user is interested in.

[0126] Through the above method, the user's preferences are analyzed and judged according to the user's browsing choices, and the user's specific browsing preferences are determined according to the user's browsing situation, so as to accurately push more content that the user is truly interested in, reduce the situation of inaccurate push, improve the user's browsing experience, enable users to browse more content they like, attract more users, and enable more users to retain for a long time.

[0127] Embodiment 2: This embodiment is an improvement made on the basis of embodiment 1. The intelligent push method based on user habits and the information analysis strategy are specifically as follows:

[0128] Get the target user's current browsing content;

[0129] Get the target user's operation on the currently browsed content;

[0130] If the target user's operation on the current browsing content is a browsing operation, then the browsing time of the target user on the current browsing content is obtained and determined as the user browsing time;

[0131] Get the total duration of the current browsing content and set it as the content duration;

[0132] Calculate the user browsing ratio, user browsing ratio = user browsing time ÷ content time;

[0133] Setting a duration ratio threshold, wherein the duration threshold is 80%;

[0134] If the user browsing ratio is ≥ the duration ratio threshold, the target adjustment strategy is executed;

[0135] If the user browsing ratio is less than the duration ratio threshold, no judgment is made, that is, the pushed content in the user's subsequent browsing process is not adjusted;

[0136] If the target user's operation on the currently browsed content is a delete operation, the user preference determination strategy is executed.

[0137] The target adjustment strategy is specifically as follows:

[0138] Get the current browsing content;

[0139] Obtain all browsing content in the database, including text, pictures and videos, and set them as target browsing content;

[0140] If there is a target browsing content identical to the current browsing content in the database, the target browsing content is identified as the target push content;

[0141] If the target browsing content identical to the current browsing content does not exist in the database, then the content keyword of the current browsing content is obtained;

[0142] Extract all target browsing contents containing content keywords from the database and identify them as target push contents;

[0143] Integrate all target push content into a push collection;

[0144] Extract any target push content in the push collection as the next current browsing content of the target user.

[0145] Embodiment 3: This embodiment is an improvement made on the basis of Embodiment 2. In this embodiment, the user preference determination strategy is specifically as follows:

[0146] Obtaining target data of the currently browsed content, wherein the target data includes title, author and content body, wherein the content body is the main content of the currently browsed content, such as characters and events;

[0147] Get all browsed content in the database;

[0148] Get the title of the currently browsed content and set it as the target title;

[0149] The browsing contents in the database with the same title as the target title are defined as the target judgment contents;

[0150] If there is target determination content in the database, all target determination content is extracted to form a target determination data set;

[0151] If the target judgment content does not exist in the database, the keywords of the target title are extracted and determined as the judgment keywords;

[0152] Extract all browsing contents containing judgment keywords in the database, identify them as target judgment contents, and form a target judgment data set;

[0153] Set the number of pushes to be determined, the number of pushes to be determined is 5;

[0154] All target judgment contents in the target judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes;

[0155] Obtain the target user's operation on each target determination content and define it as the target determination operation;

[0156] Extracting the number of target determination contents whose target determination operation is a deletion operation, and setting it as a first determination number;

[0157] If the first judgment number ÷ the judgment push number ≥ 60%, the target title and browsing content related to the judgment keyword will no longer be pushed to the target user;

[0158] Mark all target judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database;

[0159] Extract any browsing content in the push database as the next current browsing content of the target user;

[0160] If the first determination number ÷ the determined push number < 60%, the user determination strategy is executed.

[0161] This embodiment also provides that the user determination strategy is specifically:

[0162] Get all browsed content in the database;

[0163] Get the author ID of the currently browsed content and set it as the target ID;

[0164] The browsing content associated with the target ID in the database is defined as the associated determination content;

[0165] If there is association determination content in the database, all the association determination content is extracted to form an association determination data set;

[0166] If the associated determination content does not exist in the database, executing the first content determination strategy;

[0167] Get the number of push notifications;

[0168] All the related judgment contents in the related judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes;

[0169] Obtaining the target user's operation on each relevance determination content, and defining it as a relevance determination operation;

[0170] Extracting the number of associated determination contents whose associated determination operation is a deletion operation, and setting it as a second determination number;

[0171] If the second determined number ÷ the determined push number ≥ 60%, the browsing content associated with the target ID will no longer be pushed to the target user;

[0172] Mark all relevant judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database;

[0173] Extract any browsing content in the push database as the next current browsing content of the target user;

[0174] If the second determined number ÷ the determined push number < 60%, the first content determination strategy is executed.

[0175] This embodiment also provides that the first content determination strategy is specifically:

[0176] Get all browsed content in the database;

[0177] Acquire the content body of the currently browsed content, wherein the content body includes a first body, and the first body is the cover of the browsed content;

[0178] Obtain the first main body of the currently browsed content and set it as the first target content;

[0179] The browsing content having the first target content in the database is determined as the first determination content;

[0180] If the first determination content exists in the database, extract all the first determination content to form a first determination data set;

[0181] If the first content determination strategy does not exist in the database, the second content determination strategy is executed;

[0182] Get the number of push notifications;

[0183] Push all the first judgment contents in the first judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0184] Obtaining an operation of the target user on each first determination content as a first operation;

[0185] Extracting the number of first determination contents whose first operation is a deletion operation and setting it as a third determination number;

[0186] If the third determined number ÷ the determined push number ≥ 60%, the browsing content containing the first target content will no longer be pushed to the target user;

[0187] Marking all first determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0188] Extract any browsing content from the push database as the next current browsing content of the target user;

[0189] If the third determination number ÷ the determination push number < 60%, the second content determination strategy is executed.

[0190] This embodiment also provides that the second content determination strategy is specifically:

[0191] Get all browsed content in the database;

[0192] Acquire the content body of the currently browsed content, wherein the content body includes a second body, and the second body is a main character or event of the browsed content;

[0193] Obtain the second main body of the currently browsed content and set it as the second target content;

[0194] The browsing content having the second target content in the database is determined as the second determination content;

[0195] If the second determination content exists in the database, extract all the second determination content to form a second determination data set;

[0196] If the second determination content does not exist in the database, the third content determination strategy is executed;

[0197] Get the number of push notifications;

[0198] Push all the second judgment contents in the second judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0199] Obtaining an operation of the target user on each second determination content as a second operation;

[0200] Extracting the number of second determination contents in which the second operation is a deletion operation, and setting the number as a fourth determination number;

[0201] If the fourth determination number ÷ the determination push number ≥ 60%, the browsing content containing the second target content will no longer be pushed to the target user;

[0202] Marking all second determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0203] Extract any browsing content in the push database as the next current browsing content of the target user;

[0204] If the fourth determination number ÷ the determined push number < 60%, the third content determination strategy is executed.

[0205] This embodiment also provides that the third content determination strategy is specifically:

[0206] Get all browsed content in the database;

[0207] Acquire the content body of the currently browsed content, wherein the content body includes a third body, and the third body is a comment on the browsed content;

[0208] Obtain the third subject of the currently browsed content and define it as the third target content;

[0209] The browsing content having the third target content in the database is defined as the third determination content;

[0210] If the third determination content exists in the database, extract all the third determination content to form a third determination data set;

[0211] If the third determination content does not exist in the database, it is determined that the user has made an accidental touch operation, and any browsing content in the database is extracted as the next current browsing content of the target user;

[0212] Get the number of push notifications;

[0213] Push all the third judgment contents in the third judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number;

[0214] Obtaining an operation of the target user on each third determination content as a third operation;

[0215] Extracting the number of third determination contents whose third operation is a deletion operation, and setting it as a fifth determination number;

[0216] If the fifth determination number ÷ the determination push number ≥ 60%, the browsing content containing the third target content will no longer be pushed to the target user;

[0217] Marking all third-determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database;

[0218] Extract any browsing content in the push database as the next current browsing content of the target user;

[0219] If the fifth determination number ÷ the determination push number < 60%, it is determined that the user has made an accidental touch operation, and a browsing content in the database is randomly extracted as the next current browsing content of the target user.

[0220] Embodiment 4: This embodiment also discloses a system for executing a smart push method based on user habits, see Figure 2 ,include:

[0221] Acquisition module: obtain target information of target users;

[0222] Analysis module: Based on the target information, through information analysis strategy, analysis data is formed to determine the user's preferences for browsing content, so as to facilitate more accurate push of browsing content that the target user is interested in;

[0223] Judgment module: Based on the analysis data, through the target adjustment strategy or user preference judgment strategy, push data is formed to determine the preferences of the target user, and for the browsing content with deletion operation, the reason why the user performs the operation is further determined, so as to facilitate more accurate push of the browsing content that the target user is interested in;

[0224] Push module: extracts the corresponding content from the database based on the push data, and pushes the content to the target user, so as to more accurately push the browsing content that the target user is interested in.

[0225] In this embodiment, based on the user's browsing choices, the user's preferences are analyzed and judged, and based on the user's browsing situation, the user's specific browsing preferences are determined, so as to accurately push more content that the user is truly interested in, reduce the situation of inaccurate push, enhance the user's browsing experience, enable the user to browse more content they like, attract more users, and enable more users to retain for a long time.

Claims

1. An intelligent push method based on user habits, characterized by: include: Obtain target information of target users; The target information includes the target user's browsing content and operations on the browsing content; The operations include browsing operations and deleting operations; Based on the target information, the analytical data is formed through information analysis strategies; According to the analysis data, push data is generated by adjusting the strategy through goals or determining the strategy based on user preferences; According to the push data, extract the corresponding content in the database and push the content to the target users; The information analysis strategy is specifically as follows: Get the target user's current browsing content; Get the target user's operation on the currently browsed content; If the target user's operation on the current browsing content is a browsing operation, then the browsing time of the target user on the current browsing content is obtained and determined as the user browsing time; Get the total duration of the current browsing content and set it as the content duration; Calculate the user browsing ratio, user browsing ratio = user browsing time ÷ content time; Set the duration ratio threshold; If the user browsing ratio is ≥ the duration ratio threshold, the target adjustment strategy is executed; If the user browsing ratio is less than the duration ratio threshold, no judgment is made; If the target user's operation on the currently browsed content is a delete operation, the user preference determination strategy is executed.

2. The intelligent push method based on user habits according to claim 1, characterized in that: The target adjustment strategy is specifically as follows: Get the current browsing content; Obtain all browsing contents in the database and set them as target browsing contents; If there is a target browsing content identical to the current browsing content in the database, the target browsing content is identified as the target push content; If the target browsing content identical to the current browsing content does not exist in the database, then the content keyword of the current browsing content is obtained; Extract all target browsing contents containing content keywords from the database and identify them as target push contents; Integrate all target push content into a push collection; Extract any target push content in the push collection as the next current browsing content of the target user.

3. The intelligent push method based on user habits according to claim 1, characterized in that: The user preference determination strategy is specifically as follows: Obtaining target data of the currently browsed content, wherein the target data includes title, author and content body; Get all browsed content in the database; Get the title of the currently browsed content and set it as the target title; The browsing contents in the database with the same title as the target title are defined as the target judgment contents; If target determination content exists in the database, all target determination content is extracted to form a target determination data set; If the target judgment content does not exist in the database, the keywords of the target title are extracted and determined as the judgment keywords; Extract all browsing contents containing judgment keywords in the database, identify them as target judgment contents, and form a target judgment data set; Set the number of push notifications; All target judgment contents in the target judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes; Obtain the target user's operation on each target determination content and define it as the target determination operation; Extracting the number of target determination contents whose target determination operation is a deletion operation, and setting it as a first determination number; If the first judgment number ÷ the judgment push number ≥ 60%, the target title and browsing content related to the judgment keyword will no longer be pushed to the target user; Mark all target judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database; Extract any browsing content in the push database as the next current browsing content of the target user; If the first determination number ÷ the determined push number < 60%, the user determination strategy is executed.

4. The intelligent push method based on user habits according to claim 3, characterized in that: The user determination strategy is specifically: Get all browsed content in the database; Get the author ID of the currently browsed content and set it as the target ID; The browsing content associated with the target ID in the database is defined as the associated determination content; If there is association determination content in the database, all the association determination content is extracted to form an association determination data set; If the associated determination content does not exist in the database, executing the first content determination strategy; Get the number of push notifications; All the related judgment contents in the related judgment data set are pushed to the target person in order from large to small according to the number of views, and the number of pushes is the number of judgment pushes; Obtaining the target user's operation on each relevance determination content, and defining it as a relevance determination operation; Extracting the number of associated determination contents whose associated determination operation is a deletion operation, and setting it as a second determination number; If the second determined number ÷ the determined push number ≥ 60%, the browsing content associated with the target ID will no longer be pushed to the target user; Mark all relevant judgment contents in the database, and organize all unmarked browsing contents in the database to form a push database; Extract any browsing content in the push database as the next current browsing content of the target user; If the second determined number ÷ the determined push number < 60%, the first content determination strategy is executed.

5. The intelligent push method based on user habits according to claim 4, characterized in that: The first content determination strategy is specifically: Get all browsed content in the database; Acquire the content body of the currently browsed content, wherein the content body includes a first body; Obtain the first main body of the currently browsed content and set it as the first target content; The browsing content having the first target content in the database is determined as the first determination content; If the first determination content exists in the database, extract all the first determination content to form a first determination data set; If the first content determination strategy does not exist in the database, the second content determination strategy is executed; Get the number of push notifications; Push all the first judgment contents in the first judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number; Obtaining an operation of the target user on each first determination content as a first operation; Extracting the number of first determination contents whose first operation is a deletion operation and setting it as a third determination number; If the third determined number ÷ the determined push number ≥ 60%, the browsing content containing the first target content will no longer be pushed to the target user; Marking all first determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database; Extract any browsing content from the push database as the next current browsing content of the target user; If the third determination number ÷ the determination push number < 60%, the second content determination strategy is executed.

6. The intelligent push method based on user habits according to claim 5, characterized in that: The second content determination strategy is specifically: Get all browsed content in the database; Acquire the content body of the currently browsed content, wherein the content body includes a second body; Obtain the second main body of the currently browsed content and set it as the second target content; The browsing content having the second target content in the database is determined as the second determination content; If the second determination content exists in the database, extract all the second determination content to form a second determination data set; If the second determination content does not exist in the database, the third content determination strategy is executed; Get the number of push notifications. Push all the second judgment contents in the second judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number; Obtaining an operation of the target user on each second determination content as a second operation; Extracting the number of second determination contents in which the second operation is a deletion operation, and setting the number as a fourth determination number; If the fourth determination number ÷ the determination push number ≥ 60%, the browsing content containing the second target content will no longer be pushed to the target user; Marking all second determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database; Extract any browsing content in the push database as the next current browsing content of the target user; If the fourth determination number ÷ the determined push number < 60%, the third content determination strategy is executed.

7. The intelligent push method based on user habits according to claim 6, characterized in that: The third content determination strategy is specifically: Get all browsed content in the database; Acquire the content body of the currently browsed content, wherein the content body includes a third body; Obtain the third subject of the currently browsed content and define it as the third target content; The browsing content having the third target content in the database is defined as the third determination content; If the third determination content exists in the database, extract all the third determination content to form a third determination data set; If the third determination content does not exist in the database, it is determined that the user has made an accidental touch operation, and any browsing content in the database is extracted as the next current browsing content of the target user; Get the number of push notifications. Push all the third judgment contents in the third judgment data set to the target person in order from large to small according to the values ​​of the page views, and the number of pushes is the determined push number; Obtaining an operation of the target user on each third determination content as a third operation; Extracting the number of third determination contents whose third operation is a deletion operation, and setting it as a fifth determination number; If the fifth determination number ÷ the determination push number ≥ 60%, the browsing content containing the third target content will no longer be pushed to the target user; Marking all third-determined contents in the database, and organizing all unmarked browsing contents in the database to form a push database; Extract any browsing content in the push database as the next current browsing content of the target user; If the fifth determination number ÷ the determination push number < 60%, it is determined that the user has made an accidental touch operation, and a browsing content in the database is randomly extracted as the next current browsing content of the target user.

8. A system for executing the intelligent push method based on user habits as claimed in claim 1, characterized in that: include: Acquisition module: obtain target information of target users; Analysis module: Based on the target information, the analysis data is formed through information analysis strategy; Judgment module: Based on the analysis data, the strategy is adjusted through the target or the user's preference. Push module: extracts the corresponding content from the database based on the push data and pushes the content to the target users.

Images