Information recommendation method and related device

Abstract

The embodiment of the invention discloses an information recommendation method, which is used for improving the consistency and stability of recommended contents. The information recommendation method comprises the following steps: analyzing multiple types of historical behavior data of a user to obtain a data analysis result; generating a recommendation rule according to the data analysis result; sorting to-be-recommended contents based on the recommendation rule, and generating and storing a recommendation list; and receiving an access instruction sent by the user, and sending the recommendation list to the user.

Description

Technical Field

[0001] The present application relates to the field of information processing, and in particular to an information recommendation method and related devices. Background Art

[0002] With the popularization of the Internet and smart devices, personalized recommendation systems have been widely used in various digital platforms (such as e-commerce, video streaming, social media, etc.). Users can be divided into different groups based on user behavior, active time, smart tags, and device attributes, making push notifications more accurate.

[0003] Existing personalized recommendation systems are usually based on complex machine learning algorithms, which provide users with personalized content recommendations by analyzing users' historical behaviors, interest tags, and other user profile data. However, in the existing technology, the recommendation mechanism is still not comprehensive enough, resulting in a lack of consistency and stability in the recommended content during continuous access. Therefore, how to provide a personalized recommendation system to improve the consistency and stability of recommended content to enhance the overall user experience is a very worthy issue. Summary of the invention

[0004] The embodiments of the present application provide an information recommendation method and related devices for improving the consistency and stability of recommended content to enhance the overall user experience.

[0005] The first aspect of an embodiment of the present application provides an information recommendation method, comprising: analyzing multiple categories of historical behavior data of users to obtain data analysis results; generating recommendation rules based on the data analysis results; sorting the recommended content based on the recommendation rules, generating and saving a recommendation list; receiving an access instruction sent by the user, and sending the recommendation list to the user.

[0006] Optionally, in a possible embodiment, the analyzing the historical behavior data of the user to obtain the data analysis results includes: classifying and counting the videos corresponding to the multiple categories of historical behavior data by video type, and obtaining the number of valid behaviors of each video type corresponding to each category of historical behavior data; according to the weights corresponding to the each category of historical behavior data and the number of valid behaviors of each video type corresponding to the each category of historical behavior data, obtaining the recommended value of each video type as the data analysis result.

[0007] Optionally, in a possible embodiment, generating recommendation rules according to the data analysis results includes: sorting the video types in descending order of recommendation value to obtain a priority recommendation sequence; and obtaining the recommendation rules according to the priority recommendation sequence.

[0008] Optionally, in a possible embodiment, before sorting the video types in descending order of recommendation value to obtain a priority recommendation sequence as the recommendation rule, the method also includes: extracting corresponding keywords according to the video types in the videos corresponding to the multiple categories of historical behavior data; determining the target keywords corresponding to the video types, the target keywords being keywords that appear more than a first preset value or that have a probability of appearing greater than a second preset value.

[0009] Optionally, in a possible embodiment, obtaining the recommendation rule based on the priority recommendation sequence includes: arranging the target keywords corresponding to the various video types in order of number of occurrences or probability of occurrence from large to small, to obtain a sub-priority recommendation sequence corresponding to the various video types; and obtaining the recommendation rule based on the sub-priority recommendation sequence and the priority recommendation sequence.

[0010] Optionally, in a possible embodiment, the method further includes: recording the user's feedback action on the recommendation list; and updating the recommendation rule according to the feedback action.

[0011] Optionally, in a possible embodiment, the method further includes: recording and storing the user's behavior data in real time.

[0012] The second aspect of the present application provides a recommendation device, including: an analysis unit, used to analyze multiple types of historical behavior data of users to obtain data analysis results; a generation unit, used to generate recommendation rules based on the data analysis results; the generation unit is also used to sort the recommended content based on the recommendation rules, generate and save a recommendation list; a transceiver unit, used to receive an access instruction sent by the user and send the recommendation list to the user.

[0013] A third aspect of the present application provides an electronic device, comprising a memory and a processor, wherein the processor is used to implement the steps of the information recommendation method as described in any one of the first aspects above when executing a computer management program stored in the memory.

[0014] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the information recommendation processing methods described in the above aspects.

[0015] A fifth aspect of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the methods described in the above aspects.

[0016] From the above technical solution, it can be seen that the embodiment of the present application has the following advantages: by collecting and analyzing the historical behavior data of the user on the terminal device, the data analysis results are obtained, and the recommendation rules are generated according to the data analysis results to sort the recommended content, and the sorting results are saved for subsequent recommendations. This recommendation mechanism can not only reduce the real-time computing burden of the system, but also ensure the consistency and continuity of the recommended content, thereby improving the overall user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a possible information recommendation method provided in an embodiment of the present application;

[0018] Figure 1a A possible information acquisition flow chart provided for an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a possible recommended device provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of the hardware structure of another possible electronic device provided in an embodiment of the present application;

[0021] Figure 4 A schematic diagram of a possible hardware structure of a computer-readable storage medium provided in an embodiment of the present application;

[0022] Figure 5 A possible system structure diagram provided for an embodiment of the present application. DETAILED DESCRIPTION

[0023] For ease of understanding, the relevant technologies involved in the embodiments of the present application are first described below.

[0024] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0025] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the previous and next associated objects are in an "or" relationship. It should be noted that the naming of the parameters in this article is for the convenience of description. Other names can be used in practice, and this application does not limit it.

[0026] In existing personalized recommendation schemes, the ranking of recommended content usually relies on online real-time computing, which not only increases the computational burden of the system, but may also cause response delays when users obtain content. In addition, most existing recommendation systems are based on users' immediate behavior and fail to make full use of users' historical behavior data for long-term rule formulation, resulting in a lack of consistency and stability in recommended content during continuous access, which reduces user experience satisfaction. At the same time, existing recommendation systems mainly rely on real-time data processing and dynamic algorithm adjustment, which have high requirements for computing resources and are prone to recommendation lag or inaccuracy when processing large-scale user data. Specifically, there are the following problems: (1) High computing resource consumption: Existing personalized recommendation systems usually rely on complex algorithms and real-time computing to generate recommendation results. This method requires a large amount of computing resources, especially when facing massive users and diverse data, which can easily lead to excessive system load, increase hardware costs and maintenance difficulties. (2) Recommendation lag and response delay: Since personalized recommendation systems need to analyze and process users' latest behavior data in real time, there may be computing delays when users request content, resulting in a longer response time for recommendation results. This lag not only affects user experience, but also may reduce user satisfaction with recommended content. (3) Lack of consistency in recommended content: Existing recommendation systems often focus on users’ immediate behavior and ignore the analysis of long-term behavioral data. This limitation makes the recommended content inconsistent and lacks coherence across multiple visits, which in turn affects the user experience and trust in the platform. (4) Insufficient use of long-term user behavior: Although existing systems can adjust recommendation results in real time, they do not make sufficient use of users’ long-term behavioral data and fail to form effective recommendation rules. This results in the lack of pertinence and depth of recommended content in the long-term dimension, and fails to fully utilize the value of user historical data.

[0027] In view of this, in order to solve these problems, this application proposes some solutions to make recommendations based on user historical behavior data. This solution collects and analyzes the user's behavior data on the terminal device, sorts the content according to pre-established rules, and saves the sorting results for subsequent recommendations. This rule-based recommendation mechanism can not only reduce the real-time computing burden of the system, but also ensure the consistency and continuity of the recommended content, thereby improving the overall user experience.

[0028] Specifically, see Figure 1 , is a flow chart of a possible information recommendation method provided in an embodiment of the present application, comprising the following steps:

[0029] 101. Analyze multiple types of historical behavior data of users to obtain data analysis results;

[0030] The recommendation device receives a video access request sent by the user through the client through a designated port. The video access request carries at least a user ID and a video ID of the target video, wherein both the user ID and the video ID are unique. While the user is watching the video, the recommendation device collects and records the user's behavior data in real time and saves it. The behavior data includes but is not limited to at least one of the following behaviors: click behavior, viewing behavior, search behavior, and comment behavior. The collection of user behavior data includes: the terminal (such as a TV, mobile phone, tablet, etc.) records the user's click, browse, and view behavior data in real time. After collecting the user's historical behavior data, it is uploaded to a central server or cloud and stored in a database.

[0031] For the convenience of description, the behavioral data that has been generated by the user is called historical behavior data. In this application, based on the analysis of the user's historical behavior data, more accurate recommendations are made to the user, that is, the stored user behavior data is analyzed to extract the user's interest preferences and behavior patterns. Identify the differences between users' long-term and short-term behaviors. It should be noted that in order to improve the accuracy of recommendations and reduce the system storage burden, the historical behavior data can be historical data within a preset period of time. The preset period can be set according to the actual situation, such as 1 month or 15 days, such as the user's historical behavior data within 1 month.

[0032] The historical behavior data is classified according to different behaviors, including but not limited to click behaviors, viewing behaviors, search behaviors, and comment behaviors, and the user's multiple categories of historical behavior data are analyzed to obtain data analysis results. Specifically, the videos corresponding to the multiple categories of historical behavior data are classified and counted by video type, and the number of valid behaviors of each video type corresponding to each category of historical behavior data is obtained. That is, the videos corresponding to each category of historical behavior data are classified and counted according to the video type. For ease of understanding, for example, the videos corresponding to the historical click behaviors are divided into sports, entertainment, military, and life categories. Statistics show that the historical click behaviors for sports videos are 54 times, the historical click behaviors for entertainment videos are 187 times, the historical click behaviors for military videos are 9 times, and the historical click behaviors for life videos are 76 times. Similarly, the historical viewing behaviors for sports videos are 47 times, and so on. Optionally, in the present application, the number of valid behaviors can also be set. For example, in the historical viewing behavior, when the viewing time exceeds the preset time, it is a valid viewing behavior, and if it does not exceed the preset time, it is an invalid viewing behavior and will not be counted. For example, the preset time can be set to 10 seconds or 1 minute, etc., which is not limited here.

[0033] After obtaining the number of valid behaviors of each video type corresponding to each type of historical behavior data, the recommended value of each video type is obtained according to the weight corresponding to each type of historical behavior data and the number of valid behaviors of each video type corresponding to each type of historical behavior data, as the data analysis result. For example, the weight of click behavior is preset to be 0.25, the weight of viewing behavior is 0.40, the weight of search behavior is 0.65, and the data of comment behavior is 0.55. If the number of historical click behaviors of entertainment videos is 20 times, the number of historical viewing behaviors is 15 times, the number of historical search behaviors is 2 times, and the number of historical comment behaviors is 6 times, then the recommended value of entertainment videos can be calculated by the following formula: 0.25*20+0.40*15+0.65*2+0.55*6=15.6. Similarly, the recommended values ​​of other video types are calculated, and the data analysis result is obtained according to the recommended values ​​of each video type. In the present application, the recommendation value of each video type can be directly used as the data analysis result, or the data analysis result can be obtained based on the recommendation value of each video type. For example, the higher the recommendation value of the video type, the higher the corresponding recommendation level, such as level A. That is, the data analysis result can be a specific value or a specific level, etc., which is not limited in the present application.

[0034] 102. Generate recommendation rules based on data analysis results;

[0035] After obtaining the data analysis results, a recommendation rule is generated based on the data analysis results. Specifically, the video types are sorted in descending order according to the recommendation value to obtain a priority recommendation sequence. Then, based on the priority recommendation sequence, the recommendation rule is obtained. Optionally, the priority recommendation sequence can be used as a recommendation rule.

[0036] It should be noted that, considering that when users watch videos, keywords are the most likely to reflect their interests, in order to make more precise recommendations to improve the user's viewing experience, in the embodiment of the present application, before generating recommendation rules, in the videos corresponding to the multiple types of historical behavior data, the corresponding keywords are extracted according to the types of videos. It can be understood that the keyword extraction technology is a prior art, and the specific application will not be repeated. For ease of understanding, the keywords extracted from entertainment videos are Zhang San, Li Si, a certain entertainment program, etc. After extracting the keywords corresponding to each video type, the target keywords corresponding to each video type are determined, and the target keywords are keywords with a number of occurrences greater than a first preset value or a probability of occurrence greater than a second preset value. Optionally, when there are too many target keywords, or when there is a limited number, the number of occurrences or the probability of occurrence of the target keywords can also be sorted in order from large to small, and the first N target keywords are selected as the final keywords. After determining the target keywords corresponding to each video type, the target keywords of each video type are arranged in order from large to small according to the number of occurrences or the probability of occurrence, and the sub-priority recommendation sequence corresponding to each video type is obtained, and then the recommendation rule is obtained according to the sub-priority recommendation sequence and the priority recommendation sequence. It should be noted that there are many ways to obtain the recommendation rule according to the sub-priority recommendation sequence and the priority recommendation sequence, including sorting the sub-priority recommendation sequence in the order of the recommendation value of the video type, that is, the priority recommendation sequence, to obtain the recommendation rule. For example, the priority recommendation sequence is sports video>entertainment video>military video, and the sub-priority recommendation sequences corresponding to each video type are: the sub-priority recommendation sequence of sports video is A1>A2>A3, the sub-priority recommendation sequence of sports video is B1>B2>B3, and the sub-priority recommendation sequence of sports video is C1>C2>C3, among which A1-A3, B1-B3 and C1-C3 are all target keywords corresponding to each video type, and the recommendation rule obtained is:

[0037] A1>A2>A3>B1>B2>B3>C1>C2>C3. Optionally, the first M in the recommendation rule can be the first in each sub-priority recommendation sequence, and the rest can be randomly sorted or sequentially sorted, for example, the recommendation rule is: A1>B1>C1>B3>C2>B3>A2. This application does not limit this.

[0038] It should be noted that the recommendation rules are variable and adjustable to reflect the user's content type preferences, access frequency, etc.

[0039] 103. Sort the recommended contents based on the recommendation rules, generate a recommendation list and save it;

[0040] After obtaining the recommendation rules, the recommended contents are sorted based on the recommendation rules, a recommendation list is generated, and the list is saved to the cloud or server.

[0041] The sorted rules and recommendation lists are stored in the database. In this application, high-performance hbase storage or other storage methods can be used so that they can be directly used when the user visits next time.

[0042] 104. Receive an access instruction sent by the user, and send a recommendation list to the user.

[0043] The system receives the access instruction sent by the user through the terminal again, and sends the recommendation list to the user in response to the access instruction, such as displaying it on the display screen. That is, when the user visits again, the system quickly retrieves the sorted recommendation list according to the saved recommendation rules and pushes it to the user.

[0044] See also Figure 1a , which is a possible information acquisition flow chart provided in an embodiment of the present application, wherein the terminal may include a television, a mobile phone or other terminals such as a tablet, and the background service obtains the corresponding data of each device in the multi-terminal devices. For example, the mobile phone background service obtains mobile phone data, such as the mobile phone user's identification, etc. Each background service requests the unified rule recommendation service to obtain data through an RPC request. The unified rule recommendation service queries the rule recommendation data from the hbase storage according to the rowkey, and sends the corresponding rule recommendation data to each background service respectively.

[0045] Optionally, in order to improve the accuracy of the recommendation rules, the user's feedback actions on the recommendation list can also be recorded; and the recommendation rules can be updated according to the feedback actions. Specifically, the user's response to the recommended content (such as click, skip, like, etc.) is recorded, and these feedback data are used for the next rule adjustment and optimization. For example, if the user clicks on a video that is not the best recommendation in the recommendation list, the above calculation is re-performed based on the action; or if the user does not click on the recommendation list, but re-searches or clicks on other videos, the recommendation rules are updated in real time according to the feedback.

[0046] In summary, (1) by collecting and analyzing user historical behavior data, recommendation rules based on user long-term behavior are formulated, reducing dependence on real-time computing, thereby significantly reducing the system's computing resource consumption. (2) By generating and saving the sorted recommendation rules in advance, users can quickly obtain recommendation results when requesting content, avoiding response delays caused by real-time computing and improving the system's response speed. (3) By utilizing users' long-term behavior data, this application ensures the continuity and consistency of recommended content, and can provide content that meets their long-term interests and needs even during multiple visits, thereby enhancing the user experience. (4) By using rule-based recommendation methods, users' historical behavior data is fully mined and converted into valuable recommendation rules, making recommended content more accurate and effective, and improving users' stickiness and trust in the platform.

[0047] Therefore, the solution of this application collects and analyzes various behavior data of users on terminal devices, sorts the content according to pre-defined rules, and saves the sorting results for subsequent recommendations. This rule-based recommendation mechanism can not only reduce the real-time computing burden of the system, but also ensure the consistency and continuity of the recommended content, thereby improving the overall user experience.

[0048] It should be pointed out that this application has a wide range of applicability, and can be used not only in personalized recommendation systems, but also in e-commerce, streaming media, news, advertising, smart home, online education and other fields. In these occasions, this solution can significantly improve system efficiency, response speed and user experience by pre-generating rules.

[0049] The above describes the embodiment of the present application from the perspective of the information recommendation method, and the following describes the embodiment of the present application from the perspective of the recommendation device.

[0050] See also Figure 2 , Figure 2 A schematic diagram of a possible recommendation device provided in an embodiment of the present application, wherein the recommendation device specifically includes:

[0051] The analysis unit 201 is used to analyze multiple types of historical behavior data of the user to obtain data analysis results;

[0052] A generating unit 202, configured to generate a recommendation rule according to the data analysis result;

[0053] The generating unit 202 is further used to sort the recommended contents based on the recommendation rules, generate a recommendation list and save it;

[0054] The transceiver unit 203 is configured to receive an access instruction sent by the user and send the recommendation list to the user.

[0055] In a possible embodiment, the analysis unit 201 is specifically used to: classify and count the videos corresponding to the multiple categories of historical behavior data by video type, and obtain the number of valid behaviors of each video type corresponding to each category of historical behavior data; according to the weights corresponding to the various categories of historical behavior data and the number of valid behaviors of each video type corresponding to the various categories of historical behavior data, obtain the recommended value of each video type as the data analysis result.

[0056] In a possible embodiment, the generating unit 202 is specifically configured to: sort the video types in descending order according to the recommendation values ​​to obtain a priority recommendation sequence; and obtain the recommendation rule according to the priority recommendation sequence.

[0057] In a possible embodiment, the analysis unit 201 is also used to: extract corresponding keywords according to each video type in the videos corresponding to the multiple categories of historical behavior data; determine the target keywords corresponding to each video type, and the target keywords are keywords that appear more than a first preset value or have a probability of appearing greater than a second preset value.

[0058] In a possible embodiment, the generation unit 202 is specifically used to: arrange the target keywords corresponding to each video type in order of number of occurrences or probability of occurrence from large to small, to obtain a sub-priority recommendation sequence corresponding to each video type; and obtain the recommendation rule based on the sub-priority recommendation sequence and the priority recommendation sequence.

[0059] In a possible embodiment, the recommendation device further includes a recording unit 204 and an updating unit 205, wherein the recording unit 204 is specifically configured to record the user's feedback action on the recommendation list, and the updating unit 205 is specifically configured to update the recommendation rule according to the feedback action.

[0060] In a possible embodiment, the recording unit is also used to record and store the user's behavior data in real time.

[0061] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of an electronic device provided in an embodiment of the present application.

[0062] like Figure 3 As shown, an embodiment of the present application provides an electronic device, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented: analyzing multiple categories of historical behavior data of the user to obtain data analysis results; generating recommendation rules based on the data analysis results; sorting the recommended content based on the recommendation rules, generating and saving a recommendation list; receiving an access instruction sent by the user, and sending the recommendation list to the user.

[0063] Optionally, in a possible embodiment, the processor 320 is specifically used to: classify and count the videos corresponding to the multiple categories of historical behavior data by video type, and obtain the number of valid behaviors of each video type corresponding to each category of historical behavior data; according to the weights corresponding to the each category of historical behavior data and the number of valid behaviors of each video type corresponding to the each category of historical behavior data, obtain the recommended value of each video type as the data analysis result.

[0064] Optionally, in a possible embodiment, the processor 320 is specifically used to: sort the video types in descending order according to the recommendation value to obtain a priority recommendation sequence; and obtain the recommendation rule according to the priority recommendation sequence.

[0065] Optionally, in a possible embodiment, the processor 320 is also used to: extract corresponding keywords according to each video type in the videos corresponding to the multiple categories of historical behavior data; determine the target keywords corresponding to each video type, the target keywords being keywords that appear more than a first preset value or have a probability of appearing greater than a second preset value.

[0066] Optionally, in a possible embodiment, the processor 320 is specifically used to: arrange the target keywords corresponding to the video types in order of number of occurrences or probability of occurrence from large to small, to obtain a sub-priority recommendation sequence corresponding to the video types; and obtain the recommendation rule according to the sub-priority recommendation sequence and the priority recommendation sequence.

[0067] Optionally, in a possible embodiment, the processor 320 is further configured to record the user's feedback action on the recommendation list, and update the recommendation rule according to the feedback action.

[0068] Optionally, in a possible embodiment, the processor 320 is also used to record and store the user's behavior data in real time.

[0069] Since the electronic device introduced in this embodiment is a device used to implement a processing device in the embodiment of the present application, based on the method introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation method of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present application is within the scope of protection of this application.

[0070] See also Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present application.

[0071] like Figure 4As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented: analyzing multiple categories of historical behavior data of the user to obtain data analysis results; generating recommendation rules based on the data analysis results; sorting the recommended content based on the recommendation rules, generating and saving a recommendation list; receiving an access instruction sent by the user, and sending the recommendation list to the user.

[0072] The embodiment of the present application also provides a system, which may include one or more central processors 822 and memory 832, one or more storage media 830 (e.g., one or more mass storage devices) storing application programs 842 or data 844. Among them, the memory 832 and the storage medium 830 may be short-term storage or permanent storage. The program stored in the storage medium 830 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the system. Further, the central processor 822 may be configured to communicate with the storage medium 830 and execute a series of instruction operations in the storage medium 830 on the system 800. The system 800 may also include one or more power supplies 826, one or more wired or wireless network interfaces 850, one or more input and output interfaces 858, and / or, one or more operating systems 841, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.

[0073] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0074] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and 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 may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may 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 may be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may 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 (e.g., a solid state disk (SSD)), etc.

[0075] 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.

[0076] 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 an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0077] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0078] In addition, each functional unit in the embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0079] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable 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 method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0080] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An information recommendation method, characterized in that: include: Analyze multiple types of historical behavior data of users to obtain data analysis results; Generate recommendation rules based on the data analysis results; Sort the recommended contents based on the recommendation rules, generate a recommendation list and save it; An access instruction sent by the user is received, and the recommendation list is sent to the user.

2. The method according to claim 1, characterized in that The analysis of the user's historical behavior data to obtain data analysis results includes: Classifying and counting the videos corresponding to the multiple categories of historical behavior data by video type, and obtaining the number of valid behaviors of each video type corresponding to each category of historical behavior data; According to the weights corresponding to the various types of historical behavior data and the number of valid behaviors of each video type corresponding to the various types of historical behavior data, the recommended value of each video type is obtained as the data analysis result.

3. The method according to claim 2, characterized in that Generating a recommendation rule according to the data analysis result includes: Sorting the video types in descending order according to the recommendation values ​​to obtain a priority recommendation sequence; The recommendation rule is obtained according to the priority recommendation sequence.

4. The method according to claim 3, characterized in that Before sorting the video types in descending order according to the recommendation values ​​to obtain a priority recommendation sequence as the recommendation rule, the method further includes: In the videos corresponding to the multiple types of historical behavior data, extract corresponding keywords according to the types of the videos; Determine a target keyword corresponding to each video type, wherein the target keyword is a keyword whose occurrence frequency is greater than a first preset value or whose occurrence probability is greater than a second preset value.

5. The method according to claim 4, characterized in that The obtaining of the recommendation rule according to the priority recommendation sequence comprises: Arrange the target keywords corresponding to the video types in descending order according to the number of occurrences or the probability of occurrence, to obtain a sub-priority recommendation sequence corresponding to the video types; The recommendation rule is obtained according to the sub-priority recommendation sequence and the priority recommendation sequence.

6. The method according to claim 1, characterized in that The method further comprises: Recording the user's feedback action on the recommendation list; The recommendation rule is updated according to the feedback action.

7. The method according to claim 1, characterized in that The method further comprises: analyzing the user's various historical behavior data to obtain data analysis results; The user's behavior data is recorded and stored in real time.

8. A recommendation device, characterized in that: include: An analysis unit, used to analyze multiple types of historical behavior data of users to obtain data analysis results; A generating unit, used for generating a recommendation rule according to the data analysis result; The generating unit is further used to sort the recommended contents based on the recommendation rules, generate a recommendation list and save it; The transceiver unit is used to receive the access instruction sent by the user and send the recommendation list to the user.

9. A computer-readable storage medium comprising instructions, which, when executed on a computer, enables the computer to execute the method according to any one of claims 1 to 7.

10. A computer program product comprising instructions, which, when executed on a computer, enables the computer to perform the method according to any one of claims 1 to 7.

Images