New media operation content recommendation system based on big data
Through big data analysis of user browsing logs, divide related and independent browsing users, targeted recommendations for new media operation content, and dynamically supervise recommendation status, solving the problem of limited recommendations in the existing technology and improving user experience and content matching.
Patent Information
- Application Number
- CN202410854496.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-28
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-06-28
AI Technical Summary
The existing new media system is too limited when recommending new media operating content, making it difficult to tap into users' potential preferences, and large-scale recommendations may cause users to browse content that they are not interested in, reducing user experience.
A new media operating content recommendation system based on big data, through user division modules and operation content recommendation modules, analyze user browsing logs, divide related browsing users and independent browsing users, recommend operating content in a targeted manner, and dynamically supervise users through recommendation control modules The browsing status is timely interrupted and poor recommendations.
It realizes accurate recommendations based on user preferences, meets the needs of different users, improves the user experience, and ensures that users only see content they are interested in.
Smart Images

Figure CN118690083B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of new media operation technology, and more specifically, to a new media operation content recommendation system based on big data. Background Art
[0002] With the rapid development of the Internet, new media operations have become one of the important means for enterprises or individuals to conduct online marketing. At present, short videos have become the preferred marketing method for new media accounts. After an enterprise or individual registers a new media account on a new media platform, it will publish short videos of the corresponding operation content positioning type to meet the online marketing of the new media account. Users can browse the short video operation content of the new media account on the new media platform. At present, the new media system will recommend corresponding new media operation content to users based on search records and browsing records. However, this recommendation method is too limited, resulting in users only being recommended to the positioning type operation content within the record, and it is difficult to tap the user's potential preferences. At the same time, large-scale recommendations will also cause users to often browse operation content that they are not interested in, resulting in a decrease in the user's browsing experience. Summary of the invention
[0003] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a new media operation content recommendation system based on big data.
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] New media operation content recommendation system based on big data, including user segmentation module, operation content recommendation module, and recommendation management and control module;
[0006] The user classification module obtains the user's operation content browsing log and divides the user into associated browsing users or independent browsing users;
[0007] The operation content recommendation module is used to periodically recommend operation content to associated browsing users or independent browsing users;
[0008] The recommendation control module is used to obtain the user's recommendation acceptance value, and determine whether it is necessary to stop recommending operating content based on the comparison result of the recommendation acceptance value and its threshold.
[0009] Furthermore, the user's operational content browsing log is obtained. Specifically, when the user browses the operational content, an operational content browsing log is generated. The operational content browsing log includes the log number, operational content positioning type, browsing start time, browsing end time, and total duration of the operational content.
[0010] Further, the users are divided into associated browsing users or independent browsing users, specifically: obtain all operational content browsing logs of the users within n days before the current time of the system, obtain the operational content positioning type of the operational content browsing logs, mark the operational content browsing logs of the same operational content positioning type as logs of the same category, calculate the time difference between the browsing end time and the browsing start time of the logs of the same category to obtain the operational content browsing time, calculate the ratio of the operational content browsing time to the total operational content time to obtain the operational content browsing value, sum up and average all operational content browsing values of the logs of the same category to obtain the average operational content browsing value, and mark it as PED, obtain the total number of consecutive browsing times FSC, mark the total number of logs of the same category as KMA, and use the formula Get the positioning preference value RTH of the positioning type of the operation content, where a1 is the average operation content browsing value coefficient, a2 is the total number of consecutive browsing coefficient, and a3 is the total number of logs of the same type coefficient;
[0011] Setting a positioning preference threshold, when the positioning preference value of the operation content positioning type is greater than or equal to the positioning preference threshold, marking the operation content positioning type as the user's preferred positioning type;
[0012] The associated browsing value of the user is obtained, and an associated browsing threshold is set. When the associated browsing value of the user is greater than or equal to the associated browsing threshold, the user is marked as an associated browsing user. When the associated browsing value of the user is less than the associated browsing threshold, the user is marked as an independent browsing user.
[0013] Furthermore, the total number of consecutive browsing times is obtained in the following manner: sort all logs of the same category in order of log numbers, calculate the difference between the log numbers of two adjacent logs of the same category after sorting, and obtain the difference in the same category numbers; when the difference in the same category numbers is equal to 1, increase the number of consecutive browsing times by one, sum the number of consecutive browsing times of logs of the same category, obtain the total number of consecutive browsing times, and mark it as FSC.
[0014] Furthermore, the user's associated browsing value is obtained in the following way: obtain all the user's favorite positioning types, mark the total number of favorite positioning types as ZPJ, obtain the knowledge graph of the operation content positioning type, perform a pairwise association analysis on all the user's favorite positioning types, and when there is an association between two favorite positioning types, increase the number of associated recommendations by one, sum the number of associated recommendations for the user, obtain the total number of associated recommendations, and mark it as KGR, and use the formula The associated browsing value YZK of the user is obtained, wherein b1 is the total number coefficient of the preferred positioning type, and b2 is the total number coefficient of the associated recommendation.
[0015] Furthermore, the operation content is periodically recommended to the associated browsing user or the independent browsing user, specifically: when the user is marked as an associated browsing user, the preference positioning type with the largest value of the associated browsing user is marked as the associated preferred type, the knowledge graph of the operation content positioning type is obtained, the remaining operation content positioning types associated with the associated preferred type are marked as associated recommended types, all the preference positioning types of the associated browsing user are marked as associated deleted types, all the associated deleted types in the associated recommended types are deleted, and based on the preset cycle, the operation content of the remaining associated recommended types is recommended to the associated browsing user at each cycle node;
[0016] When a user is marked as an independent browsing user, the preference positioning type with the largest positioning preference value is marked as the independent preferred type of the independent browsing user, and the remaining independent browsing users with the same independent preferred type as the independent browsing user are marked as reference independent browsing users, and the preference positioning type of the reference independent browsing user is obtained, and the independent browsing recommendation value of each preference positioning type is obtained, and an independent browsing recommendation threshold is set. When the independent browsing recommendation value of the preference positioning type is greater than or equal to the independent browsing recommendation threshold, the preference positioning type is marked as an independent recommendation type, and all preference positioning types of the independent browsing user are marked as recommended deletion types. All recommended deletion types in the independent recommendation types are deleted, and based on the preset cycle, the operational content of the remaining independent recommendation types is recommended to the independent browsing user at each cycle node.
[0017] Furthermore, the exclusive browsing recommendation value of the preference positioning type is obtained in the following manner: marking the reference exclusive browsing users with the same preference positioning type as the same-recommended reference exclusive browsing users, summing up the total number of the same-recommended reference exclusive browsing users, and obtaining the exclusive browsing recommendation value of the preference positioning type.
[0018] Furthermore, the user's recommendation acceptance value is obtained, and based on the comparison result between the recommendation acceptance value and its threshold, it is determined whether it is necessary to stop recommending operational content. Specifically, after the system recommends operational content to the user, each time the user generates an operational content browsing log, the user's recommendation acceptance value is obtained, and the recommendation acceptance threshold is set. When the recommendation acceptance value is greater than or equal to the recommendation acceptance threshold, the operational content recommended to the user in the current cycle is canceled, and operational content of the type of preference positioning is displayed to the user.
[0019] Furthermore, the user's recommendation acceptance value is obtained in the following manner: obtain all operational content browsing logs after the operational content is recommended to the user in the current period, obtain the operational content browsing value of the operational content browsing log, sort all operational content browsing logs in order of log numbers, mark the previous operational content browsing log adjacent to the sorting as the previous browsing log, mark the next operational content browsing log adjacent to the sorting as the next browsing log, calculate the ratio of the operational content browsing value of the next browsing log to the operational content browsing value of the previous browsing log, obtain the browsing state change value, and set the browsing state threshold change value , when the browsing state change value is less than the browsing state threshold change value, the number of browsing state declines is increased once, the number of browsing state declines is summed up to obtain the total number of browsing state declines, and marked as RWK, the operation content browsing values of all operation content browsing logs are summed up and averaged to obtain the recommended content browsing average, and marked as SDX, the time difference between the browsing start time of the post-browsing log and the browsing end time of the pre-browsing log is calculated to obtain the recommended browsing interval, all recommended browsing intervals are summed up and averaged to obtain the average recommended browsing interval, and marked as MCZ, using the formula The recommendation acceptance value ZAP is obtained, where c1 is the total number of browsing state decline coefficients, c2 is the average coefficient of recommended content browsing, and c3 is the average recommended browsing interval coefficient.
[0020] Compared with the prior art, the present invention has the following beneficial effects:
[0021] 1. Setting up a user segmentation module and an operation content recommendation module can obtain the user's preferred positioning type based on the operation content browsing log, and further analyze the user's preferred positioning type, divide the user into associated browsing users and independent browsing users, and analyze the potential needs of associated browsing users and independent browsing users in a targeted manner, and periodically recommend operation content of different positioning types to associated browsing users and independent browsing users, so as to meet the new media operation content recommendation needs of different users;
[0022] 2. Setting up a recommendation management and control module can continuously supervise the recommendation of new media operation content, dynamically analyze the user's browsing status of recommended content, and promptly interrupt the recommendation of new media operation content when the user's browsing status of recommended content is not good, so as to ensure the user's browsing experience of new media operation content. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 A flowchart for recommending operational content to associated browsing users or independent browsing users;
[0024] Figure 2 A flowchart for determining whether it is necessary to stop recommending operational content. DETAILED DESCRIPTION
[0025] Example 1
[0026] Reference Figure 1 , a new media operation content recommendation system based on big data, including user segmentation module and operation content recommendation module.
[0027] User segmentation module: When a user browses the operational content, an operational content browsing log is generated. The operational content browsing log includes a log number (the log number is i, i=1, 2, ...N, N is a positive integer, and each operational content browsing log of the same user corresponds to an independent and unique log number), an operational content positioning type (each operational content corresponds to an operational content positioning type, such as the operational content positioning type of operational content A is a food recipe sharing type, the operational content positioning type of operational content B is a food store exploration type, and the operational content positioning type of operational content C is a travel guide sharing type), browsing start time (taking short videos as an example, that is, the start time of short video playback), browsing end time (taking short videos as an example, that is, the end time of short video playback), and total duration of the operational content (such as the total duration of short video operational content).
[0028] For example: the log number of the operation content browsing log is 23, the operation content account positioning type is the food recipe sharing type, the browsing start time is 13:15:38 on February 11, 2024, the browsing end time is 13:19:40 on February 11, 2024, and the total duration of the operation content is 8 minutes.
[0029] Get all the operational content browsing logs of the user within n days before the current time of the system, get the operational content positioning type of the operational content browsing log, mark the operational content browsing logs of the same operational content positioning type as logs of the same category, sort all the logs of the same category in the order of log numbers, calculate the difference between the log numbers of two adjacent logs of the same category after sorting, and get the difference of the same category number. When the difference of the same category number is equal to 1, increase the number of consecutive browsing times by one, and when the difference of the same category number is not equal to 1, do not do the corresponding processing, calculate the time difference between the browsing end time and the browsing start time of the log of the same category, and get the browsing time of the operational content, calculate the ratio of the browsing time of the operational content to the total duration of the operational content, and get the browsing value of the operational content, sum up all the browsing values of the log of the same category and take the average, and get the average browsing value of the operational content, and mark it as PED, sum up the number of consecutive browsing times of the log of the same category, and get the total number of consecutive browsing times, and mark it as FSC, mark the total number of logs of the same category as KMA, and use the formula The positioning preference value RTH of the operational content positioning type is obtained, where a1 is the average operational content browsing value coefficient, a2 is the total number of consecutive browsing coefficient, and a3 is the total number of logs of the same type coefficient. The value of a1 is 0.97, the value of a2 is 0.99, and the value of a3 is 0.53.
[0030] Set the positioning preference threshold. The positioning preference threshold is the system setting threshold and can be modified according to actual needs.
[0031] When the positioning preference value of the operation content positioning type is greater than or equal to the positioning preference threshold, the operation content positioning type is marked as the user's preferred positioning type.
[0032] When the positioning preference value of the operational content positioning type is less than the positioning preference threshold, no corresponding processing is performed.
[0033] Get all the user's favorite positioning types, mark the total number of favorite positioning types as ZPJ, get the knowledge graph of the operation content positioning type, perform pairwise correlation analysis on all the user's favorite positioning types, and when there is a correlation between two favorite positioning types, increase the number of associated recommendations by one. When there is no correlation between two favorite positioning types, do not do the corresponding processing, sum the number of associated recommendations for the user, get the total number of associated recommendations, and mark it as KGR. Use the formula The associated browsing value YZK of the user is obtained, where b1 is the total number coefficient of the preferred positioning type, b2 is the total number coefficient of the associated recommendation, the value of b1 is 0.74, and the value of b2 is 0.92.
[0034] The operational content positioning type is represented in the form of a knowledge graph. For example, the relationship in the knowledge graph represents the association between the operational content positioning types, such as a containment relationship, a parallel relationship, etc., to ensure that the knowledge graph can accurately reflect the relationship between the operational content positioning types. For example, if there is a parallel relationship between the food exploration type and the food recipe sharing type, then the food exploration type and the food recipe sharing type are associated in the knowledge graph.
[0035] Set the associated browsing threshold. The associated browsing threshold is a system-set threshold that can be modified according to actual needs. When the user's associated browsing value is greater than or equal to the associated browsing threshold, the user is marked as an associated browsing user. When the user's associated browsing value is less than the associated browsing threshold, the user is marked as an independent browsing user.
[0036] Operational content recommendation module: When a user is marked as an associated browsing user, the preference positioning type with the largest value of the associated browsing user is marked as the associated preferred type, and the knowledge graph of the operational content positioning type is obtained. The remaining operational content positioning types associated with the associated preferred type are marked as associated recommended types, and all the preference positioning types of the associated browsing user are marked as associated deleted types. All associated deleted types in the associated recommended types are deleted, and based on the preset cycle, the operational content of the remaining associated recommended types is recommended to the associated browsing user at each cycle node.
[0037] When a user is marked as an independent browsing user, the preference positioning type with the largest positioning preference value is marked as the independent preferred type of the independent browsing user, and the remaining independent browsing users with the same independent preferred type as the independent browsing user are marked as reference independent browsing users, and the preference positioning type of the reference independent browsing user is obtained, and the independent browsing recommendation value of each preference positioning type is obtained, and an independent browsing recommendation threshold is set. When the independent browsing recommendation value of the preference positioning type is greater than or equal to the independent browsing recommendation threshold, the preference positioning type is marked as an independent recommendation type. When the independent browsing recommendation value of the preference positioning type is less than the independent browsing recommendation threshold, no corresponding processing is performed, and all preference positioning types of the independent browsing user are marked as recommended deletion types. All recommended deletion types in the independent recommendation types are deleted, and based on the preset cycle, the operating content of the remaining independent recommendation types is recommended to the independent browsing user at each cycle node.
[0038] The exclusive browsing recommendation value of the preference positioning type is obtained in the following way: marking the reference exclusive browsing users with the same preference positioning type as the same-recommended reference exclusive browsing users, summing up the total number of the same-recommended reference exclusive browsing users, and obtaining the exclusive browsing recommendation value of the preference positioning type. For example: there are 3 reference exclusive browsing users in total. The preference positioning types of reference exclusive browsing user a include travel guide sharing type and clothing matching type, the preference positioning types of reference exclusive browsing user b include travel guide sharing type and food exploration type, and the preference positioning types of reference exclusive browsing user c include travel guide sharing type and funny type. If these 3 are the same reference exclusive browsing users for the travel guide sharing type, then the unique browsing recommendation value of the travel guide sharing type is 3. Only reference exclusive browsing user a belongs to the same reference exclusive browsing user for the clothing matching type, then the unique browsing recommendation value of the clothing matching type is 1. Only reference exclusive browsing user b belongs to the same reference exclusive browsing user for the food exploration type, then the unique browsing recommendation value of the food exploration type is 1. Only reference exclusive browsing user c belongs to the same reference exclusive browsing user for the funny type, then the unique browsing recommendation value of the funny type is 1.
[0039] By setting up a user segmentation module and an operation content recommendation module, you can obtain the user's preferred positioning type based on the operation content browsing log, and further analyze the user's preferred positioning type, divide the users into associated browsing users and independent browsing users, and analyze the potential needs of associated browsing users and independent browsing users in a targeted manner, and periodically recommend operation content of different positioning types to associated browsing users and independent browsing users to meet the new media operation content recommendation needs of different users.
[0040] Example 2
[0041] Reference Figure 2 On the basis of Example 1, it also includes a recommendation management and control module. The recommendation management and control module: when the system recommends operational content to a user, each time the user generates an operational content browsing log, the recommendation acceptance value of the user is obtained, and a recommendation acceptance threshold is set. The recommendation acceptance threshold is a threshold set by the system and can be modified according to actual needs. When the recommendation acceptance value is greater than or equal to the recommendation acceptance threshold, the operational content recommended to the user in the current cycle is canceled, and the operational content of the user's preference positioning type is displayed to the user. When the recommendation acceptance value is less than the recommendation acceptance threshold, no corresponding processing is performed.
[0042] The user's recommendation acceptance value is obtained in the following way: obtain all the operation content browsing logs of the user in the current period, obtain the operation content browsing value of the operation content browsing log, sort all the operation content browsing logs in the order of log numbers, mark the previous operation content browsing log adjacent to the sorting as the previous browsing log, mark the next operation content browsing log adjacent to the sorting as the next browsing log, calculate the ratio of the operation content browsing value of the next browsing log to the operation content browsing value of the previous browsing log, obtain the browsing state change value, set the browsing state threshold change value, and when the browsing state When the variable value is less than the browsing state threshold variable value, the number of browsing state declines is increased by one, the number of browsing state declines is summed up to obtain the total number of browsing state declines, and marked as RWK, the operation content browsing values of all operation content browsing logs are summed up and averaged to obtain the recommended content browsing average, and marked as SDX, the time difference between the browsing start time of the post-browsing log and the browsing end time of the pre-browsing log is calculated to obtain the recommended browsing interval, all recommended browsing intervals are summed up and averaged to obtain the average recommended browsing interval, and marked as MCZ, using the formula The recommendation acceptance value ZAP is obtained, where c1 is the total number of browsing status decline coefficients, c2 is the recommended content browsing mean coefficient, and c3 is the average recommended browsing interval coefficient. The value of c1 is 0.82, the value of c2 is 0.91, and the value of c3 is 0.68.
[0043] By setting up a recommendation management and control module, you can continuously supervise the recommendation of new media operation content, dynamically analyze the user's browsing status of recommended content, and promptly interrupt the recommendation of new media operation content when the user's browsing status of recommended content is not good, to ensure the user's browsing experience of new media operation content.
[0044] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0045] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0046] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0047] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0048] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0049] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0050] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.
[0051] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A new media operation content recommendation system based on big data, characterized by: Including user segmentation module, operation content recommendation module, and recommendation management and control module; The user classification module obtains the user's operation content browsing log and divides the user into associated browsing users or independent browsing users; The operation content recommendation module is used to periodically recommend operation content to associated browsing users or independent browsing users; The recommendation control module is used to obtain the recommendation acceptance value of the associated browsing user or the independent browsing user, and determine whether it is necessary to stop recommending the operating content based on the comparison result between the recommendation acceptance value and its threshold value; The users are divided into associated browsing users or independent browsing users, specifically: obtain all the operational content browsing logs of the users within n days before the current time of the system, obtain the operational content positioning type of the operational content browsing logs, mark the operational content browsing logs of the same operational content positioning type as logs of the same category, calculate the time difference between the browsing end time and the browsing start time of the logs of the same category to obtain the operational content browsing time, calculate the ratio of the operational content browsing time to the total operational content time to obtain the operational content browsing value, sum up and average all the operational content browsing values of the logs of the same category to obtain the average operational content browsing value, and mark it as PED, obtain the total number of consecutive browsing times FSC, mark the total number of logs of the same category as KMA, and use the formula Get the positioning preference value RTH of the positioning type of the operation content, where a1 is the average operation content browsing value coefficient, a2 is the total number of consecutive browsing coefficient, and a3 is the total number of logs of the same type coefficient; Setting a positioning preference threshold, when the positioning preference value of the operation content positioning type is greater than or equal to the positioning preference threshold, marking the operation content positioning type as the user's preferred positioning type; Obtain the associated browsing value of the user, set the associated browsing threshold, and when the associated browsing value of the user is greater than or equal to the associated browsing threshold, mark the user as an associated browsing user; when the associated browsing value of the user is less than the associated browsing threshold, mark the user as an independent browsing user; The user's associated browsing value is obtained in the following way: obtain all the user's favorite positioning types, mark the total number of favorite positioning types as ZPJ, obtain the knowledge graph of the operation content positioning type, perform a pairwise association analysis on all the user's favorite positioning types, and when there is a correlation between two favorite positioning types, increase the number of associated recommendations by one, sum the number of associated recommendations for the user, obtain the total number of associated recommendations, and mark it as KGR. Get the user's associated browsing value YZK, where b1 is the total number coefficient of the preferred positioning type, and b2 is the total number coefficient of associated recommendations; Periodically recommending operational content to associated browsing users or independent browsing users, specifically: when a user is marked as an associated browsing user, marking the associated browsing user's largest preference positioning type as an associated preferred type, obtaining a knowledge graph of operational content positioning types, marking the remaining operational content positioning types associated with the associated preferred type as associated recommended types, marking all the associated browsing user's preference positioning types as associated deleted types, deleting all associated deleted types in the associated recommended types, and recommending the remaining associated recommended types of operational content to the associated browsing user at each cycle node based on a preset cycle; When a user is marked as an independent browsing user, the preference positioning type with the largest positioning preference value is marked as the independent preferred type of the independent browsing user, and the remaining independent browsing users with the same independent preferred type as the independent browsing user are marked as reference independent browsing users, and the preference positioning type of the reference independent browsing user is obtained, and the independent browsing recommendation value of each preference positioning type is obtained, and an independent browsing recommendation threshold is set. When the independent browsing recommendation value of the preference positioning type is greater than or equal to the independent browsing recommendation threshold, the preference positioning type is marked as an independent recommendation type, and all preference positioning types of the independent browsing user are marked as recommended deletion types. All recommended deletion types in the independent recommendation types are deleted, and based on the preset cycle, the operational content of the remaining independent recommendation types is recommended to the independent browsing user at each cycle node.
2. The new media operation content recommendation system based on big data according to claim 1 is characterized in that: Obtain the user's operational content browsing log, specifically: when the user browses the operational content, an operational content browsing log is generated. The operational content browsing log includes the log number, operational content positioning type, browsing start time, browsing end time, and total duration of the operational content.
3. The new media operation content recommendation system based on big data according to claim 1 is characterized in that: The total number of consecutive browsing times is obtained in the following way: sort all logs of the same category in the order of log numbers, calculate the difference between the log numbers of two adjacent logs of the same category after sorting, and obtain the difference in the same category numbers; when the difference in the same category numbers is equal to 1, increase the number of consecutive browsing times by one, sum the number of consecutive browsing times of logs of the same category, obtain the total number of consecutive browsing times, and mark it as FSC.
4. The new media operation content recommendation system based on big data according to claim 1 is characterized in that: The exclusive browsing recommendation value of the preference positioning type is obtained in the following way: marking the reference exclusive browsing users with the same preference positioning type as the same-recommended reference exclusive browsing users, summing up the total number of the same-recommended reference exclusive browsing users, and obtaining the exclusive browsing recommendation value of the preference positioning type.
5. The new media operation content recommendation system based on big data according to claim 1 is characterized in that: Obtain the recommendation acceptance value of the associated browsing user or the independent browsing user, and determine whether it is necessary to stop recommending the operational content based on the comparison result between the recommendation acceptance value and its threshold. Specifically: after the system recommends operational content to the user, each time the user generates an operational content browsing log, obtain the user's recommendation acceptance value and set the recommendation acceptance threshold. When the recommendation acceptance value is greater than or equal to the recommendation acceptance threshold, cancel the recommendation of operational content to the user in the current cycle, and display the operational content of the preference positioning type to the user.
6. The new media operation content recommendation system based on big data according to claim 1 is characterized in that: The user's recommendation acceptance value is obtained in the following way: obtain all the operation content browsing logs of the user in the current period, obtain the operation content browsing value of the operation content browsing log, sort all the operation content browsing logs in the order of log numbers, mark the previous operation content browsing log adjacent to the sorting as the previous browsing log, mark the next operation content browsing log adjacent to the sorting as the next browsing log, calculate the ratio of the operation content browsing value of the next browsing log to the operation content browsing value of the previous browsing log, obtain the browsing state change value, set the browsing state threshold change value, and when the browsing state When the variable value is less than the browsing state threshold variable value, the number of browsing state declines is increased by one, the number of browsing state declines is summed up to obtain the total number of browsing state declines, and marked as RWK, the operation content browsing values of all operation content browsing logs are summed up and averaged to obtain the recommended content browsing average, and marked as SDX, the time difference between the browsing start time of the post-browsing log and the browsing end time of the pre-browsing log is calculated to obtain the recommended browsing interval, all recommended browsing intervals are summed up and averaged to obtain the average recommended browsing interval, and marked as MCZ, using the formula The recommendation acceptance value ZAP is obtained, where c1 is the total number of browsing state decline coefficients, c2 is the average coefficient of recommended content browsing, and c3 is the average recommended browsing interval coefficient.
Citation Information
Patent Citations
Product intelligent recommendation method and device, computer equipment and storage medium
CN110135937A
Content similarity analysis method and device based on knowledge graph
CN112732927A
Intelligent recommendation method and system
CN117633339A
Media systems for temporally and contextually relevant recommendations
US20170195731A1
Method for providing recommended channel list, and display device according thereto
WO2020017930A1
