Cloud video recommendation method and system and storage medium

Abstract

The invention discloses a cloud video recommendation method and system and a storage medium, and relates to the technical field of cloud video recommendation, and the method comprises the steps: firstly investigating a user, then tracking the behavior of the user, laying a foundation for the confirmation of the learning style type of the subsequent user, and improving the user experience. Secondly, the adaptive learning style type of the user is determined based on the questionnaire filled by the user, the watching videos and the behavior data corresponding to the watching videos, the adaptability of the recommended learning videos and the learning style of the user is high, the learning efficiency of the user is improved, the learning interest and memory effect of the user are improved, the pertinence is high, the requirements of different users are met, and the user experience is improved. And when the learning video recommendation is performed on the user, the adaptive learning style type of the user is considered, and the function of the learning node of the user is also considered, so that the user is helped to digest and absorb the preview course, the preview effect of the user is improved, the learning video recommendation is performed on the place where the user is insufficient in learning, and the consolidation effect of the user is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of cloud video recommendation, and in particular to a cloud video recommendation method, system and storage medium. Background Art

[0002] Video platform teaching or "cloud desktop" has a clear trend of participating in school teaching. Cloud video applications are the most popular field because they do not require equipment maintenance, performance upgrades, or software updates. Cloud video transmission or remote control of cloud servers through high-speed networks is the choice of more and more universities. By allowing students to query content and learn on certain cloud video platforms, the quality of learning can be greatly improved. Therefore, it is worth studying how to use the cloud desktop video platform to provide users with video recommendations for deep learning functions to meet expectations.

[0003] Most of the existing technologies determine the videos that users are interested in based on the users' preference for the videos, and then make personalized recommendations. This method is not suitable for school education platforms, and when recommending learning videos for student users, it lacks judgment on the learning styles of student users. Different learning styles of users affect the learning efficiency of users. At the same time, it also lacks recommendations for preview videos for users' expected nodes and consolidation videos for users' consolidation nodes, which is not conducive to users' digestion and absorption of preview courses. Users are more blind when entering teaching, which is not easy for users to understand and reduces the preview effect of users. Summary of the invention

[0004] The purpose of this application is to provide a cloud video recommendation method, system and storage medium to solve the problems existing in the background technology.

[0005] In order to solve the above technical problems, the present application adopts the following technical solutions: The first aspect of the present application provides a cloud video recommendation method, including: S1. When a user logs in to the cloud video recommendation platform, a questionnaire is provided to the user, and the user's questionnaire is collected after filling it out.

[0006] S2. Track the user's behavior, record the user's watched videos, and summarize the user's watched videos and their corresponding behavior data.

[0007] S3. Based on the user's questionnaire, combined with the user's viewing videos and their corresponding behavior data, determine the user's adaptive learning style type, where the learning style types include visual, auditory and hands-on.

[0008] S4. According to the user's adaptive learning style type, combined with the user's learning subjects and the videos watched and their corresponding behavioral data, the user's exploration recommended learning videos and consolidation recommended learning videos are screened.

[0009] S5. The user's exploration recommended learning videos and consolidation recommended learning videos are displayed on the user's visual display terminal.

[0010] The second aspect of the present application provides a system for executing the cloud video recommendation method described in the present application, including: a user survey module, which is used to provide a questionnaire to the user when the user logs in to the cloud video recommendation platform, and collect the user's questionnaire after filling it out.

[0011] The behavior tracking module is used to track the user's behavior, record the user's watched videos, and summarize the user's watched videos and their corresponding behavior data.

[0012] The learning style type determination module is used to determine the user's adaptive learning style type based on the user's questionnaire, combined with the user's viewing videos and their corresponding behavior data, where the learning style types include visual, auditory and hands-on.

[0013] The learning video recommendation module is used to screen the user's exploration recommended learning videos and consolidation recommended learning videos according to the user's adaptive learning style type, combined with the user's learning subjects and the videos watched and their corresponding behavioral data.

[0014] The display terminal is used to display the user's various exploration recommended learning videos and various consolidation recommended learning videos on the user's visual display terminal.

[0015] A third aspect of the present application provides a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the cloud video recommendation method described in the present application.

[0016] The beneficial effects of the present application are: (1) the present application firstly surveys the users and then tracks the users' behaviors, thus laying a foundation for the subsequent confirmation of the users' learning style types.

[0017] (2) This application determines the user's adaptive learning style type based on the questionnaire filled out by the user and the videos watched and their corresponding behavioral data, which makes up for the lack of prior art in this regard. The recommended learning videos are highly compatible with the user's learning style, improve the user's learning efficiency, increase the user's learning interest and memory effect, and are highly targeted to meet the needs of different users.

[0018] (3) When recommending learning videos to users, this application not only takes into account the user's adaptive learning style type, but also takes into account the role of the user's learning nodes, which helps the user digest and absorb the preview courses, making the user more proficient when entering the teaching, facilitating the user's understanding, and improving the user's preview effect. Learning videos are recommended for the user's learning deficiencies to ensure the user's consolidation effect in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 This is a schematic diagram of the structure connection of a cloud video recommendation system according to an embodiment of the present application.

[0021] Figure 2 The present invention is a flowchart of the steps for implementing a cloud video recommendation method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only 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.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the application pertains. The terms used herein in the specification are for the purpose of describing specific embodiments only and are not intended to limit the present application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0024] It should be noted that in the following description, the terms "first", "second" and "third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first", "second" and "third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application can be implemented in an order other than that shown or described.

[0025] Embodiment 1

[0026] A cloud video recommendation platform (or "cloud desktop") is deployed for users, and accounts are distributed to users for accessing remote videos.

[0027] Reference Figure 1 As shown, the first aspect of the present application provides a cloud video recommendation method, including: S1. When a user logs in to a cloud video recommendation platform, a questionnaire is provided to the user, and the user's questionnaire is collected after filling it out.

[0028] S2. Track the user's behavior, record the user's watched videos, and summarize the user's watched videos and their corresponding behavior data.

[0029] In a specific embodiment of the present application, the behavior data includes the viewing time of each key image, the listening time of each target audio segment, and the download status characteristic value of each exercise question, and the download status includes downloaded and not downloaded.

[0030] It should be noted that if the download status is downloaded, the download status characteristic value is recorded as A, and if the download status is not downloaded, the download status characteristic value is recorded as A', wherein A>A'.

[0031] It should be noted that the key images are specifically images that the user pauses or returns to watch, the target audio segments are specifically detailed explanation audios, and the listening time of the target audio segments is the total time of the user's initial listening time and the returned listening time.

[0032] This application first surveys users and then tracks their behavior, laying the foundation for the subsequent confirmation of users' learning style types.

[0033] S3. Based on the user's questionnaire, combined with the user's viewing videos and their corresponding behavior data, determine the user's adaptive learning style type, where the learning style types include visual, auditory and hands-on.

[0034] In a specific embodiment of the present application, the method for determining the user's adaptive learning style type is as follows: based on the user's questionnaire, extract the options of each objective survey question and the reply text and reply time of each subjective survey question, and evaluate the user's first tendency x in each learning style type. 0_i , i is the number of each learning style type, i∈[1,3].

[0035] Based on the user's watched videos and their corresponding behavior data, determine the user's second tendency x in each learning style type 1_i .

[0036] Import the user's first and second tendencies of each learning style type into the overall tendency assessment model In the formula, the overall tendency of the user in each learning style type is output. If the overall tendency of a learning style type is the largest, then this learning style type is recorded as the adaptive learning style type, where δ _0 , δ _1 They are respectively expressed as the proportion factor of the first tendency and the proportion factor of the second tendency.

[0037] It should be noted that the proportion factor of the first tendency and the proportion factor of the second tendency are specifically set by the staff of the cloud video recommendation platform. The purpose is to adjust the influence of the first tendency and the second tendency on the overall tendency. For example, the analysis basis of the first tendency is a questionnaire survey, and the user's subjective consciousness is strong. The analysis basis of the second tendency is behavioral data, which is more objective. When setting the proportion factor of the first tendency and the proportion factor of the second tendency, the proportion factor of the first tendency can be smaller than the proportion factor of the second tendency.

[0038] In a specific embodiment of the present application, the first tendency of the user in each learning style type is evaluated by a specific evaluation method as follows: based on the selection items of each objective survey question of the user, combined with the target selection items of each learning style type in each objective survey question stored in the database, the objective survey questions of the user in each learning style type are screened and obtained, the total number of the user's objective survey questions and the number of objective survey questions in each learning style type are summarized, and the number of the user's objective survey questions in each learning style type is divided by the total number of objective survey questions to obtain the user's objective tendency β in each learning style type. _0i .

[0039] Based on the user's reply texts for each subjective survey question, the standard reply texts for each subjective survey question and the subjective survey questions corresponding to each learning style type are extracted from the database, the user's standard reply texts for each subjective survey question of each learning style type are screened, the correct rate of the user's subjective survey questions of each learning style type is obtained through data processing, the user's qualified subjective survey questions of each learning style type are screened, the total number of the user's subjective survey questions and the number of qualified subjective survey questions of each learning style type are counted, and the number of qualified subjective survey questions of the user in each learning style type is divided by the total number of subjective survey questions to obtain the user's subjective tendency β in each learning style type. _1i .

[0040] Import the user's objective and subjective tendencies in each learning style type into the first tendency evaluation model χ 0_i =β _0i *λ _0 +β _1i *λ _1 , output the user's first inclination in each learning style type.

[0041] Where λ _0 , _1 is the weight factor of the objective tendency and the weight factor of the subjective tendency stored in the database, where λ _0 <λ _1 , and λ _0 +λ _1 =1.

[0042] It should be noted that the correct rate of the user in each subjective survey question of each learning style type is specifically analyzed by: constructing a keyword set E of the user's reply to each subjective survey question of each learning style type according to the user's reply text to each subjective survey question of each learning style type im Similarly, we construct the standard response keyword set E for each subjective survey question of each learning style type. i ' m , through the formula The output is the correct rate of the user in each subjective survey question of each learning style type, where m is the number of each subjective survey question, m=1, 2, ..., l, and l is any integer greater than 2.

[0043] It should also be noted that the specific analysis method for screening the user's qualified subjective survey questions in each learning style type is: comparing the accuracy of the user's subjective survey questions in each learning style type with the accuracy threshold stored in the database; if the accuracy of a subjective survey question is greater than or equal to the accuracy threshold, then the subjective survey question is recorded as a qualified survey question, and the user's qualified subjective survey questions in each learning style type are obtained.

[0044] It should be understood that the accuracy threshold is specifically set by the staff according to specific circumstances, for example, in order to improve the analysis accuracy of the user's learning style type, the accuracy threshold is increased.

[0045] In a specific embodiment of the present application, the second tendency of the user in each learning style type is determined by the following specific method: based on each video watched by the user, each target audio of each video watched by the user is screened, each learning style type of each video watched by the user is determined, and each video watched by the user in each learning style type is mapped.

[0046] It should be noted that the specific screening method for screening the target audios of the videos watched by the user is as follows: based on the videos watched by the user, the explanation audios of the videos watched by the user are extracted, thereby obtaining several keywords in the audios of the videos watched by the user.

[0047] The number of keywords in several audios of each video watched by the user is summarized and compared with the keyword number threshold of the detailed explanation audio stored in the database. If the keyword number of a certain audio is greater than or equal to the keyword number threshold of the detailed explanation audio, the audio is recorded as the target audio, and the target audios of each video watched by the user are screened.

[0048] Get the viewing time T of each key image of each video watched by the user 1bp , the listening time of each target audio segment U2bj and the download status eigenvalue R of each exercise 3bh , b is the number of each watched video, b=1,2,...,d, d is any integer greater than 2, p is the number of each key image, p=1,2,...,q, q is any integer greater than 2, j is the number of each target audio segment, j=1,2,...,k, k is any integer greater than 2, h is the number of each exercise, h=1,2,...,g, g is any integer greater than 2.

[0049] Count the total number of videos watched by the user and the number of videos watched by each learning style type, and import it together with the user's behavior data in each video watched into the second tendency evaluation model , output the second inclination of the user in each learning style type.

[0050] Where M1, M2, and M3 are the number of videos watched by users with visual learning style, the number of videos watched by users with auditory learning style, and the number of videos watched by users with hands-on learning style, respectively. bp 、U2′ bj They respectively represent the preset viewing time of the pth key image in the bth viewing video and the preset listening time of the jth target audio stored in the database, and q, k, and g are the number of key images, the number of target audios, and the number of exercises, respectively.

[0051] In a specific embodiment of the present application, the learning style type of each video watched by the user is determined by the following specific determination method: each video watched by the user is divided according to the number of video frames, a number of sub-pictures of each video watched by the user are obtained, and the sub-pictures are compared with each other. If a sub-picture is the same as another sub-picture, the sub-pictures are grouped into a picture set, and then a number of picture sets of each video watched by the user are obtained, and the number of picture sets of each video watched by the user is counted. _1b .

[0052] Based on each target audio of each video watched by the user, the number of target audios of each video watched by the user is counted I _2b .

[0053] Based on each video watched by the user, obtain each exercise question of each video watched by the user, and count the number of exercises of each video watched by the user I _3b .

[0054] If I _1b ≥I′ _1 , then the learning style type of watching videos is recorded as visual type, I′ _1 The number of visual learning style image sets stored in the database.

[0055] If I _2b ≥I′ _2 , then the learning style type of watching videos is recorded as auditory type, I′ _2 The number of target audio files with auditory learning style type stored in the database.

[0056] If I _3b ≥I′ _3 , then the learning style type of watching videos is recorded as hands-on type, I′ _3 The number of exercises with a hands-on learning style type stored in the database.

[0057] The learning style types of each video watched by the user are obtained by summarizing.

[0058] This application determines the user's adaptive learning style type based on the questionnaire filled out by the user and the videos watched and their corresponding behavioral data, which makes up for the lack of neglect in this aspect in the prior art. The recommended learning videos are highly adaptable to the user's learning style, improve the user's learning efficiency, increase the user's learning interest and memory effect, and are highly targeted to meet the needs of different users.

[0059] S4. According to the user's adaptive learning style type, combined with the user's learning subjects and the videos watched and their corresponding behavioral data, the user's exploration recommended learning videos and consolidation recommended learning videos are screened.

[0060] In a specific embodiment of the present application, the screening of the user's exploration recommended learning videos and consolidation recommended learning videos is specifically determined by: based on the user's adaptive learning style type, obtaining from the database the exploration learning videos corresponding to each expected learning node in each learning style type, obtaining the expected learning node corresponding to the user from the cloud video recommendation platform, screening to obtain the user's expected learning node in the adaptive learning style type of the exploration recommended videos, and recording them as the user's exploration recommended learning videos.

[0061] It should be noted that the expected learning node corresponding to the user is uploaded by the teacher corresponding to the user.

[0062] The test data corresponding to the user is obtained from the cloud video recommendation platform, where the test data includes the learning mastery of each historical learning node, and combined with the user's behavioral data in each video watched, the user's various consolidation recommended learning videos are determined.

[0063] It should be noted that the user's learning mastery at each historical learning node is specifically comprehensively judged by the user's online exercise accuracy rate and offline class quality coefficient at each historical learning node. Specifically, they can be averaged to obtain the user's learning mastery at each historical learning node, or the online exercise accuracy rate and offline class quality coefficient can be weighted, and the allocated online exercise accuracy rate and offline class quality coefficient can be averaged to obtain the user's learning mastery at each historical learning node. The offline class quality coefficient is specifically judged and uploaded by the user's corresponding teacher.

[0064] In a specific embodiment of the present application, the specific determination method for determining the user's various consolidation recommended learning videos is: comparing the user's learning mastery at each historical learning node with a preset learning mastery threshold; if the user's learning mastery at a certain historical learning node is less than the learning mastery threshold, then the historical learning node is recorded as a consolidation learning node, and the user's various consolidation learning nodes are obtained by screening.

[0065] Based on the user's watched videos and historical learning nodes, combined with the consolidation learning videos of each learning node in each learning style type stored in the database, the user's watched videos at each historical learning node are filtered, and the user's behavioral data of each watched video at each historical learning node are obtained to filter the user's consolidation learning nodes.

[0066] It should be noted that the specific screening method for screening the user's consolidation learning nodes is: the viewing time of each key image of each video watched by the user at each historical learning node, the listening time of each target audio segment and the download status characteristic value of each exercise question, if the viewing time of each key image of a certain video watched is greater than or equal to the preset viewing time, the listening time of each target audio segment is greater than or equal to the preset listening time or the download status characteristic value of each exercise question is A, then the watched video is recorded as a valid watched video, and the number of valid watched videos and the total number of watched videos are counted, and the number of valid watched videos is divided by the total number of watched videos to obtain the efficiency of the user at each historical learning node, and compare it with the preset efficiency of each historical learning node stored in the database. If the efficiency of a certain historical node is less than the preset efficiency, then the historical node is recorded as a consolidation learning node to obtain the user's consolidation learning nodes.

[0067] It should also be noted that the preset effectiveness of each historical learning node is set and uploaded by the teacher.

[0068] The user's various consolidation learning nodes are summarized, and based on the user's adaptive learning style type and the various consolidation learning videos of each learning node in each learning style type, the user's various consolidation learning nodes in the adaptive learning style type are screened to obtain the various consolidation learning videos, which are recorded as the user's various consolidation recommended learning videos.

[0069] When recommending learning videos to users, this application not only takes into account the user's adaptive learning style type, but also takes into account the role of the user's learning nodes, which helps users digest and absorb the preview courses, making them more proficient when entering the teaching, facilitating user understanding, and improving the user's preview effect. Learning videos are recommended for the user's learning deficiencies to ensure the user's consolidation effect in a timely manner.

[0070] S5. The user's exploration recommended learning videos and consolidation recommended learning videos are displayed on the user's visual display terminal.

[0071] Embodiment 2

[0072] Reference Figure 2 As shown, the second aspect of the present application is a system for the cloud video recommendation method described in the present application, comprising: a user survey module, a behavior tracking module, a learning style type determination module, a learning video recommendation module and a display terminal.

[0073] It should be noted that the present application also includes a database for storing target selection items for each learning style type in each objective survey question, standard response text for each subjective survey question, each subjective survey question corresponding to each learning style type, weight factors of objective tendencies and weight factors of subjective tendencies, preset viewing time of each key picture in each viewing video, preset listening time of each target audio segment, the number of picture sets with visual learning style type, the number of target audio with auditory learning style type, the number of exercises with hands-on learning style type, each exploratory learning video corresponding to each learning style type, each expected learning node in the exploratory learning video corresponding to each learning style type, each consolidation learning video for each learning node in each learning style type, accuracy threshold, keyword number threshold for detailed explanation audio, and preset efficiency of each historical learning node.

[0074] It should also be noted that the user survey module is connected to the behavior tracking module, the behavior tracking module is connected to the learning style type determination module, the learning style type determination module is connected to the learning video recommendation module, the learning video recommendation module is connected to the display terminal, and the database is connected to the learning style type determination module and the learning video recommendation module respectively.

[0075] The user survey module is used to provide a questionnaire to the user when the user logs in to the cloud video recommendation platform, and collect the user's questionnaire after filling it out.

[0076] The behavior tracking module is used to track the user's behavior, record the user's watched videos, and summarize the user's watched videos and their corresponding behavior data.

[0077] The learning style type determination module is used to determine the user's adaptive learning style type based on the user's questionnaire and in combination with the user's watched videos and their corresponding behavior data, wherein the learning style types include visual, auditory and hands-on.

[0078] The learning video recommendation module is used to screen the user's exploration recommended learning videos and consolidation recommended learning videos according to the user's adapted learning style type, combined with the user's learning subjects and the videos watched and their corresponding behavior data.

[0079] The display terminal is used to display the user's various exploration recommended learning videos and various consolidation recommended learning videos on the user's visual display terminal.

[0080] Embodiment 3

[0081] A third aspect of the present application provides a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the cloud video recommendation method described in the present application.

[0082] Embodiment 4

[0083] In the video and computing power cooperation project between the applicant and a well-known fashion design university, the technology applied for was widely used in teachers' scientific research and teaching, and students' research and learning, and achieved good results.

[0084] This application is further explained in conjunction with the applicant's cooperative project with a well-known fashion design college.

[0085] Pre-install cloud desktop software as interface software on the original old equipment of the school or the mobile terminal devices of teachers and students, and distribute accounts and passwords to each registered user. Account functions are set differently, such as teachers have corresponding management permissions and student accounts are ordinary accounts.

[0086] New student users are required to fill out a user questionnaire when logging in and their information will be collected. The content of the user questionnaire is determined by the responsible teacher. The video records watched and browsed by student users in their daily learning will be summarized by the system background to obtain behavioral data, including the viewing time of each key image, the listening time of each target audio segment, and the download status characteristic value of each exercise question, etc. The data can be used for teachers' reference. Based on the questionnaire filled out by student users and combined with the videos they watched and their corresponding behavioral data, the system automatically identifies or the teacher manually marks the student users as visual, auditory, and hands-on.

[0087] Based on the learning style type of each student user determined above, combined with their learning subjects and the videos they watch and their corresponding behavior data, the user's exploration recommended learning videos and consolidation recommended learning videos can be screened. Teachers can also discover and record popular teaching courses for uploading based on this.

[0088] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0089] The above contents are merely examples and explanations of the inventive concept of the present application. The technical personnel in the relevant technical field may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the inventive concept of the application or exceed the scope defined in the present application, they should all fall within the protection scope of the present application.

Claims

1. A cloud video recommendation method, characterized in that: include: S1. When a user logs in to the cloud video recommendation platform, provide the user with a questionnaire and collect the user's questionnaire after filling it out; S2. Track user behavior, record user's watched videos, and summarize the user's watched videos and their corresponding behavior data; S3. Based on the user's questionnaire, combined with the user's watched videos and their corresponding behavior data, determine the user's adaptive learning style type, where the learning style types include visual, auditory and hands-on; S4. According to the user's adaptive learning style type, combined with the user's learning subjects and the videos watched and their corresponding behavior data, the user's exploration recommended learning videos and consolidation recommended learning videos are screened; S5. The user's exploration recommended learning videos and consolidation recommended learning videos are displayed on the user's visual display terminal.

2. A cloud video recommendation method according to claim 1, characterized in that: The behavior data includes the viewing time of each key image, the listening time of each target audio segment, and the download status characteristic value of each exercise question, and the download status includes downloaded and not downloaded.

3. A cloud video recommendation method according to claim 2, characterized in that: The specific determination method of determining the user's adapted learning style type is as follows: Based on the user's questionnaire, extract the options of each objective survey question and the response text and response time of each subjective survey question, and evaluate the user's first tendency in each learning style type. 0_i , i is the number of each learning style type, i∈[1,3]; Based on the user's watched videos and their corresponding behavior data, determine the user's second tendency x in each learning style type 1_i ; Import the user's first and second tendencies of each learning style type into the overall tendency assessment model In the formula, the overall inclination of the user in each learning style type is output. If the overall inclination of a certain learning style type is the largest, then this learning style type is recorded as the adaptive learning style type, where δ _0 , δ _1 They are respectively expressed as the proportion factor of the first tendency and the proportion factor of the second tendency.

4. A cloud video recommendation method according to claim 3, characterized in that: The specific evaluation method of evaluating the first inclination of the user in each learning style type is as follows: Based on the selection items of each objective survey question of the user, combined with the target selection items of each learning style type in each objective survey question stored in the database, the objective survey questions of each learning style type of the user are screened, and the total number of objective survey questions of the user and the number of objective survey questions of each learning style type are summarized to obtain the objective tendency β of the user in each learning style type. _0i ; Based on the user's reply texts for each subjective survey question, the standard reply texts for each subjective survey question and the subjective survey questions corresponding to each learning style type are extracted from the database, the user's standard reply texts for each subjective survey question of each learning style type are screened, the correct rate of the user's subjective survey questions of each learning style type is obtained through data processing, the user's qualified subjective survey questions of each learning style type are screened, the total number of the user's subjective survey questions and the number of qualified subjective survey questions of each learning style type are counted, and the number of qualified subjective survey questions of the user in each learning style type is divided by the total number of subjective survey questions to obtain the user's subjective tendency β in each learning style type. _1i ; Import the user's objective and subjective tendencies in each learning style type into the first tendency evaluation model χ 0_i =β _0i *λ _0 +β _1i *λ _1 , output the user's first inclination in each learning style type; Where λ _0 , _1 is the weight factor of the objective tendency and the weight factor of the subjective tendency stored in the database, where λ _0 <λ _1 , and λ _0 +λ _1 =1.

5. A cloud video recommendation method according to claim 2, characterized in that: The specific determination method of determining the second inclination of the user in each learning style type is as follows: Based on the videos watched by the user, each target audio of each video watched by the user is screened, each learning style type of each video watched by the user is determined, and each video watched by the user in each learning style type is mapped; Get the viewing time T of each key image of each video watched by the user 1bp , the listening time of each target audio segment U 2bj and the download status eigenvalue R of each exercise 3bh , b is the number of each video to be watched, b = 1, 2, ..., d, d is any integer greater than 2, p is the number of each key image, p = 1, 2, ..., q, q is any integer greater than 2, j is the number of each target audio segment, j = 1, 2, ..., k, k is any integer greater than 2, h is the number of each exercise, h = 1, 2, ..., g, g is any integer greater than 2; Count the total number of videos watched by the user and the number of videos watched by each learning style type, and import it together with the user's behavior data in each video watched into the second tendency evaluation model , outputting the second inclination of the user in each learning style type; Where M1, M2, and M3 are the number of videos watched by users with visual learning style, the number of videos watched by users with auditory learning style, and the number of videos watched by users with hands-on learning style, respectively. bp 、U2′ bj They respectively represent the preset viewing time of the pth key image in the bth viewing video and the preset listening time of the jth target audio stored in the database, and q, k, and g are the number of key images, the number of target audios, and the number of exercises, respectively.

6. A cloud video recommendation method according to claim 5, characterized in that: The specific determination method of determining the learning style types of each video watched by the user is as follows: Divide each video watched by the user according to the number of video frames, obtain several sub-pictures of each video watched by the user, and compare them with each other. If a sub-picture is the same as other sub-pictures, they are summarized into a picture set, and then several picture sets of each video watched by the user are obtained, and the number of picture sets of each video watched by the user is counted. _1b ; Based on each target audio of each video watched by the user, the number of target audios of each video watched by the user is counted I _2b ; Based on each video watched by the user, obtain each exercise question of each video watched by the user, and count the number of exercises of each video watched by the user I _3b ; If I _1b ≥I′ _1 , then the learning style type of watching videos is recorded as visual type, I′ _1 is the number of image sets with visual learning style type stored in the database; If I _2b ≥I′ _2 , then the learning style type of watching videos is recorded as auditory type, I′ _2 The number of target audios whose learning style type is auditory type stored in the database; If I _3b ≥I′ _3 , then the learning style type of watching videos is recorded as hands-on type, I′ _3 The number of exercises with hands-on learning style type stored in the database; The learning style types of each video watched by the user are obtained by summarizing.

7. A cloud video recommendation method according to claim 1, characterized in that: The specific method for determining the exploration recommended learning videos and consolidation recommended learning videos of the screening users is as follows: Based on the user's adaptive learning style type, the exploration learning videos corresponding to each expected learning node in each learning style type are obtained from the database, the expected learning node corresponding to the user is obtained from the cloud video recommendation platform, and the exploration recommended videos of the user's expected learning node in the adaptive learning style type are screened and recorded as the user's exploration recommended learning videos; The test data corresponding to the user is obtained from the cloud video recommendation platform, where the test data includes the learning mastery of each historical learning node, and combined with the user's behavioral data in each video watched, the user's various consolidation recommended learning videos are determined.

8. A cloud video recommendation method according to claim 7, characterized in that: The specific method for determining each user's recommended consolidation learning video is as follows: Compare the user's learning mastery at each historical learning node with the preset learning mastery threshold. If the user's learning mastery at a certain historical learning node is less than the learning mastery threshold, then record the historical learning node as a consolidation learning node, and filter out the user's consolidation learning nodes; Based on the user's watched videos and historical learning nodes, combined with the consolidation learning videos of each learning node in each learning style type stored in the database, the user's watched videos at each historical learning node are screened, and the user's behavior data of each watched video at each historical learning node are obtained, so as to screen the user's consolidation learning nodes; The user's various consolidation learning nodes are summarized, and based on the user's adaptive learning style type and the various consolidation learning videos of each learning node in each learning style type, the user's various consolidation learning nodes in the adaptive learning style type are screened to obtain the various consolidation learning videos, which are recorded as the user's various consolidation recommended learning videos.

9. A system for executing the cloud video recommendation method according to any one of claims 1 to 8, characterized in that: include: The user survey module is used to provide a questionnaire to the user when the user logs in to the cloud video recommendation platform, and collect the user's questionnaire after filling it out; The behavior tracking module is used to track the user's behavior, record the user's watched videos, and summarize the user's watched videos and their corresponding behavior data; A learning style type determination module is used to determine the user's adaptive learning style type based on the user's questionnaire, combined with the user's watched videos and their corresponding behavior data, where the learning style types include visual, auditory and hands-on; The learning video recommendation module is used to screen the user's exploration recommended learning videos and consolidation recommended learning videos according to the user's adaptive learning style type, combined with the user's learning subject and each watched video and its corresponding behavior data; The display terminal is used to display the user's various exploration recommended learning videos and various consolidation recommended learning videos on the user's visual display terminal.

10. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the cloud video recommendation method according to any one of claims 1 to 8.

Images