Online music course recommendation method and system based on data processing

Abstract

The invention belongs to the technical field of data processing, and provides an online music course recommendation method and system based on data processing, and the method comprises the steps: collecting first data of students, the first data at least comprising music interests and preferences, music levels, musical instrument ownership conditions, curriculum schedules, and historical music homework submission time; preprocessing the first data to generate a multi-dimensional user portrait, and generating a user-course interaction matrix according to the multi-dimensional user portrait; determining music practice available time of the student according to the curriculum schedule and the historical music homework submission time; according to the music interest preference, the music level, the musical instrument ownership condition and the music practice available time, the students are aggregated into a learning group; and generating a course recommendation list for each student based on the learning group and the user-course interaction matrix, and pushing the course recommendation list to the student.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to an online music course recommendation method and system based on data processing. Background Art

[0002] As part of Internet education, online music courses have developed rapidly in recent years. Unlike traditional offline music teaching, online music courses are flexible and diverse, breaking through the limitations of time and space, and providing convenient learning opportunities for students of different backgrounds and levels. These courses not only cover a variety of music styles such as classical, jazz, pop, electronic, and folk, but also include the study of a variety of instruments such as piano, guitar, violin, and vocal music, as well as composition, arrangement, music theory, recording, and mixing. The course settings cover everything from entry-level to advanced professional courses, which can meet different levels of learning needs such as basic music theory learning, performance skill improvement, and music creation.

[0003] In order to improve the user's learning experience and help students find more suitable courses, existing online music course recommendation systems usually adopt data-driven recommendation methods. These systems generate user profiles by analyzing the user's basic information, interest preferences, and historical learning behavior, and recommend personalized courses for users.

[0004] However, music courses are courses that require a lot of practice. The existing recommendation methods do not take into account the difficulties of online learning and offline practice, such as factors such as musical instruments and practice time. Mutual practice among classmates after online courses can partially solve the problem of practice difficulties, but the existing course recommendation system does not take into account the differences in levels and practice time between different students, and it is difficult for students to form mutual assistance groups. Summary of the invention

[0005] In order to solve the problems in the prior art, the present invention provides an online music course recommendation method based on data processing, the method comprising the following steps: Collecting first data of the student, the first data at least including music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time; Preprocessing the first data to generate a multi-dimensional user portrait, and generating a user-course interaction matrix according to the multi-dimensional user portrait; Determine the student's available time for music practice based on the course schedule and the time for submission of historical music assignments; Aggregating students into study groups based on the musical interest preferences, musical level, musical instrument ownership, and available time for music practice; Based on the study group and the user-course interaction matrix, a course recommendation list is generated for each student, and the course recommendation list is pushed to the student.

[0006] Another aspect of the present invention further provides an online music course recommendation system based on data processing, the system comprising the following modules: A collection module, used for collecting first data of students, wherein the first data at least includes music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time; A processing module, configured to pre-process the first data to generate a multi-dimensional user portrait, and generate a user-course interaction matrix according to the multi-dimensional user portrait; A generating module, for determining the student's available time for music practice according to the course schedule and the historical music homework submission time; An aggregation module, for aggregating students into learning groups according to the music interest preference, music level, musical instrument ownership, and available time for music practice; A push module is used to generate a course recommendation list for each student based on the study group and the user-course interaction matrix, and push the course recommendation list to the student.

[0007] The online music course recommendation method based on data processing has the following beneficial effects: By building a multi-dimensional user profile, we can fully consider the students' musical interests, musical level, instrument ownership, and available practice time to generate a personalized course recommendation list. This ensures that the recommended courses are highly matched to the students' needs, improving the pertinence and effectiveness of learning.

[0008] By grouping students with similar interests and abilities into study groups, the possibility of interaction and cooperative learning among students is enhanced. Group learning can promote knowledge sharing and skill improvement among students, improve overall learning efficiency, and enhance the fun and sustainability of learning.

[0009] When recommending courses, we fully consider the students' schedules and the time to submit their historical music homework, and ensure that the recommended courses are reasonably arranged in the students' free time to ensure the feasibility of the courses. This avoids the time conflict problem caused by unreasonable study arrangements and improves the feasibility of learning.

[0010] After the group is formed, the system will dynamically adjust the structure of the study group according to the students' learning performance and interest changes to ensure that the students in each group have a high degree of similarity in interests and learning abilities, thereby maximizing the collaboration and interaction of group learning.

[0011] In summary, this recommendation method provides students with a more personalized, collaborative, flexible and feasible music course learning solution by integrating user portraits, diversified recommendation algorithms, dynamic management of learning groups and real-time feedback mechanism, which greatly improves students' learning effects and learning experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0013] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0014] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0015] This embodiment solves the above problem through the following steps: In one embodiment, reference Figure 1 The present invention provides an online music course recommendation method based on data processing, which involves a technical solution for generating a personalized course recommendation list for a user according to factors such as the user's interest preferences, musical level, available practice time, and actual performance conditions by collecting and analyzing user-related data.

[0016] The first data of the students is collected, wherein the first data at least includes music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time.

[0017] Collecting first data of students, wherein the first data at least includes the following contents: The student's music interest preferences involve different music styles and types, including but not limited to personal preferences for classical music, jazz music, pop music, electronic music, folk music, rock music, blues music, country music, world music and other music styles, and further refine the interest preference information based on specific music categories, composers, performance forms and other parameters.

[0018] The students' musical level includes their mastery of basic music theory, specifically the degree of mastery of basic music knowledge such as notes, rhythm, mode, chords, intervals, etc. It also includes the students' performance level on specific musical instruments, such as the proficiency of playing techniques, the types and difficulty levels of the instrument playing techniques mastered, specific sight-reading ability, music score recognition ability, and the complexity of the performance works, etc. It also includes the students' ability in music creation, such as melody writing, harmony configuration, rhythm creation, music arrangement, etc.

[0019] The student's ownership of musical instruments, including whether the student owns a specific type of musical instrument, the type of musical instrument (such as piano, guitar, violin, drums, etc.), the number and specific model and brand of the musical instrument, the usage status of the musical instrument (such as the newness, whether it is used normally, maintenance status, etc.), and whether the student owns auxiliary equipment (such as pedals, speakers, amplifiers, etc.). The student's course schedule information, including the student's current full course schedule, not limited to music courses, but covers the specific arrangements of all courses in progress, including weekly class times, class periods, course names, course types, course durations, and time arrangements with other learning activities, and further including the student's free time every day. Combined with the course schedule, it is possible to determine the student's free time and free time available for music learning and practice in a day or a week, thereby providing a basis for formulating a reasonable music learning plan and course recommendations.

[0020] The student's historical music homework submission time includes the time data of various types of homework submitted by the student in the past during the music learning process, including the submission date of the homework, the name of the course involved, the type of homework (such as performance homework, music theory homework, creation homework, etc.), the completion time of the homework, the homework score and the teacher's feedback information, as well as the time regularity of the student's task completion and the timeliness of submission reflected in a specific learning stage, thereby providing a basis for subsequent learning progress planning and recommendations.

[0021] The first data is preprocessed to generate a multi-dimensional user portrait, and a user-course interaction matrix is ​​generated according to the multi-dimensional user portrait.

[0022] In this step, the first data is preprocessed to generate a multidimensional user portrait. The preprocessing process includes data cleaning, missing value filling, data standardization and normalization, and finally forms a user feature data set with high quality and consistency. On this basis, a multidimensional user portrait is further constructed, and a user-course interaction matrix is ​​generated based on the multidimensional user portrait to support the implementation of personalized recommendations.

[0023] First, the data of three dimensions, namely music interest preference, music level and instrument ownership, are preprocessed so as to better represent the user's characteristics when generating the user-course interaction matrix.

[0024] Encode the user's music preferences. Use the TF-IDF (term frequency-inverse document frequency) method to quantify the music style (such as classical, jazz, pop, etc.) to obtain a vector, in which each dimension represents the user's preference for a specific music style.

[0025] Standardized scoring of the user's music level. This score is composed of multiple features such as music theory knowledge, the complexity of performance skills, and proficiency in works, to obtain a comprehensive music level score. The score is quantified between 0 and 1 to reflect the user's current overall music level.

[0026] Use one-hot encoding to encode the types of instruments owned by the user (such as piano, guitar, violin, etc.) to form an instrument ownership vector , where each dimension indicates whether the user owns a specific musical instrument, with a value of 0 or 1.

[0027] Construct feature vectors for users and courses Based on the above three features, the user's music interest, music level, and instrument ownership are combined to generate a user feature vector ; For each course, a feature vector is generated based on the content characteristics of the course. in: is the music style vector of the course, indicating the music styles and proportions involved in the course; Score the difficulty of the course, with a numerical value between 0 and 1, reflecting the musical level required by the course; is a vector of the types of musical instruments required for the course, indicating the types of musical instruments needed to learn the course.

[0028] Content-based recommendation score calculation Content-based recommendation generates a preliminary user preference rating for a course by calculating the similarity between the user and course feature vectors.

[0029] Use cosine similarity to calculate the user's music interest vector With course music style vector The similarity between them is as follows: This value is used to measure the user's potential interest in the course style.

[0030] Music level suitability: Calculate the user's music level score Difficulty rating of the course The degree of fit between: Among them, the closer the adaptation degree is to 1, the more it means that the user's music level is more matched with the difficulty of the course.

[0031] Instrument matching: User's musical instrument ownership The instruments required for the course The matching degree between them is calculated using the vector dot product: The dot product result indicates the degree of match between the user and the instrument required for the course. The larger the value, the higher the match.

[0032] Combining the above three similarity scores, we can get the content-based rating of user 𝑢 for course 𝑐 , using the weighted average method: in, is the weight coefficient of different similarity scores, satisfying , which can be adjusted according to the actual scenario.

[0033] Furthermore, it captures the similarities between users and the commonality of their learning habits.

[0034] Calculate the similarity between users based on their music interests, musical levels, and instrument ownership.

[0035] Use the Pearson correlation coefficient to measure the similarity between users: in, is the rating of user 𝑖 on course 𝑗, is the average rating of user 𝑖, The average rating of users.

[0036] Based on the historical ratings of similar users, About Courses Rating Make predictions: This formula is used to predict the likely ratings a user would give to a course they have not yet taken.

[0037] Content-based scoring and rating prediction Fusion to generate the final user-course interaction matrix The specific formula is as follows: in, and is the fusion coefficient, satisfying The setting of the fusion coefficient can be adjusted according to the performance of the model to ensure that the recommendation results can reflect the user's content preferences and combine the experience and behavior of similar users.

[0038] Through the above steps and improvements, the generated user-course interaction matrix not only comprehensively considers the user's musical interests, levels and instrument ownership, but also ensures the accuracy, personalization and dynamic adaptability of the recommendation results to changes in user interests.

[0039] The students' available time for music practice is determined based on the course schedule and the submission time of historical music homework.

[0040] The student's available time for music practice is determined based on the course schedule and historical music homework submission time. Specifically, the course schedule data includes all the student's current course schedules, the time period, frequency and duration of each course, and the historical music homework submission time includes the date and time of past homework submissions. By combining this information, the student's study and free time are analyzed, and the available time for music practice is scientifically and reasonably allocated to provide a basis for personalized course recommendations.

[0041] To determine the time students have available for music practice, consider their class schedule and homework completion patterns. Use the following steps: Course schedule data analysis Time period division: Divide a week into multiple time periods from morning to night, usually using 1 hour as a time period (which can be adjusted to half an hour or 15 minutes as needed).

[0042] Constructing a time matrix ,in Indicates whether the student has a course in the 𝑗th time period of the 𝑖th day. If yes, ,otherwise Traverse all time periods of each day and mark all The time period is the student's free time period, and a list of all free time in a week is generated.

[0043] Historical job submission time analysis Job submission frequency statistics: Statistics are collected on the submission time of historical music homework, the time patterns of students completing homework are analyzed, and the practice time periods that students are more accustomed to are identified.

[0044] Use time series analysis to count the time points of homework submission, find out the time windows in a week when students frequently do music learning and homework, and generate a priority time period list.

[0045] Combine your free time and priorities to set practice time The intersection of free time and prioritized practice time: The best practice time for students is determined by finding the intersection of the free time in the course schedule and the priority time period of the homework. It can be expressed as: in: It means that the student’s 𝑗th time period on the 𝑖th day is an idle time period; It represents the priority time for submitting a job in the 𝑗th time period of the 𝑖th day.

[0046] Based on the list of optimal practice times generated by the intersection results, the time periods within a week are sorted according to indicators such as the length of free time and the frequency of priority time, and finally a timetable suitable for students to practice music is determined.

[0047] For example: Assume that a student's class schedule and historical music homework submission time information are as follows: Course Schedule Information There are school classes from 8:00 am to 12:00 pm from Monday to Friday.

[0048] There are additional subject tutorials on Tuesday and Thursday afternoons from 14:00 to 16:00.

[0049] The rest of the time is free time.

[0050] Historical Music Assignment Submission Date In the past month, the student habitually submitted music assignments between 20:00 and 21:00 at night, at least 3 times a week.

[0051] Specific steps to generate the available time for practice: Time matrix construction: The time period of a week is divided into 24 hours a day, one time period per hour, for a total of 168 time periods.

[0052] Construct a time matrix, where the time period from 8:00 to 12:00 from Monday to Friday is , Tuesday and Thursday from 14:00 to 16:00 , the rest of the time Idle time slot marker: From the time matrix 𝑇, we label The time period is free time.

[0053] The obtained free time periods include 12:00 to 14:00 and 16:00 to 24:00 from Monday to Friday, and all day on weekends (0:00 to 24:00).

[0054] Prioritize practice time calculation: Based on the historical music homework submission time, the analysis found that 20:00 to 21:00 every night was the students' usual practice time. Priority time periods were generated and marked in a week.

[0055] Intersection Finding: Find the intersection of the idle time period and the priority time period: Monday to Friday: The time period from 20:00 to 21:00 in the evening is free time and is also the historical priority practice time, so it is marked as the best practice time.

[0056] The whole day on Saturday and Sunday is also free time. In combination with the habit of giving priority to practice, it is recommended to arrange part of this time for concentrated practice.

[0057] Determine the time available for practice: The resulting music practice available time is: Monday to Friday evenings from 20:00 to 21:00.

[0058] All day on Saturday and Sunday, with 20:00 to 21:00 recommended as the focus practice time.

[0059] Final Practice Schedule Based on the above analysis, the available timetable for students' music practice is determined as follows: Monday to Friday: 20:00 - 21:00.

[0060] Saturday and Sunday: flexible arrangement for the whole day, and it is recommended to focus on practice from 20:00 to 21:00 in the evening.

[0061] In this way, the student's practice schedule combines the free time of his course and his historical learning habits, so as to ensure that the practice time is in line with the student's daily rhythm and has high feasibility and learning efficiency. This time allocation method can help the recommendation system provide students with more targeted music course recommendations, ensuring that the learning effect is maximized based on the feasibility of the student's practice time.

[0062] Students are grouped into study groups based on their musical interest preferences, musical level, musical instrument ownership, and available time for music practice.

[0063] In order to group students with similar interests, abilities, and schedules into study groups, a clustering algorithm is used to classify students. The specific steps include: Each student's interest preference vector , Music Level Rating , Musical Instrument Ownership Vector , practice time vector Combined to form a comprehensive feature vector , used for cluster analysis.

[0064] The similarity between the feature vectors of every two students is calculated. In order to adapt to the nature of multi-dimensional features, Euclidean distance or cosine similarity is used to measure the similarity between students, where Euclidean distance is used to calculate the absolute difference of features in different dimensions, and cosine similarity is used to measure the consistency of the direction of students' interests and time arrangements.

[0065] The K-means clustering algorithm is used to classify students. The specific steps are as follows: Determine the number of clusters K: Based on the number of students and the distribution of different music styles and levels, choose an appropriate number of clusters K so that the number of students in each group is moderate and manageable.

[0066] Randomly initialize centroids: Randomly select K students’ feature vectors as the initial cluster centers.

[0067] Iteratively calculate distance: For each student, calculate the distance between him and each cluster center, and assign the student to the cluster center with the closest distance.

[0068] Update centroids: For all students in each cluster, calculate new centroids and update the center of each cluster.

[0069] Convergence condition judgment: Repeat the iteration until the cluster center no longer changes or the change amount is lower than the set threshold. The clustering process ends and K learning groups are finally formed.

[0070] On the basis of the clustering results, the characteristic matching degree of students in the group was further checked to ensure that the students in each study group had a high degree of consistency in terms of music style interests, practice time arrangements, instrument ownership, etc., so as to ensure the synergy within the study group.

[0071] Scoring based on student's musical level , some students with higher levels can be arranged in the study group as group leaders or auxiliary coaches to help other group members learn, thereby improving the learning efficiency and cooperation ability of the entire group.

[0072] This aggregation method ensures that members of each study group have high similarities in interests, levels, musical instruments, etc., so as to maximize learning collaboration and overall learning effects.

[0073] Based on the study group and the user-course interaction matrix, a course recommendation list is generated for each student, and the course recommendation list is pushed to the student.

[0074] Based on the study group and the user-course interaction matrix, a personalized course recommendation list is generated for each student. Specifically, the information of students with similar interest preferences, similar music levels, and similar musical instruments in the study group is first summarized, and the collaborative learning potential between group members is used to analyze the common needs of each group for music courses. At the same time, combined with the historical preferences and interest matching of students for various courses recorded in the user-course interaction matrix, a score prediction between each student and all optional courses is generated.

[0075] In the process of generating the course recommendation list, the system adopts a combination of content-based recommendation algorithms, which not only recommends similar courses to students by analyzing their interest in existing courses, but also uses the learning situation of other members in the group to modify and optimize the recommendation results based on their choices and performance. The course recommendation list that each student finally obtains not only reflects his or her own interests, level, and availability of learning time, but also reflects the learning consistency within the group, aiming to improve students' learning effects and interactive experience through the group learning model.

[0076] The generated personalized course recommendation list will be sorted according to the student's learning preferences, with the courses that best meet the student's needs and are of moderate difficulty placed at the top of the list. After the recommendation is completed, the system will push the recommendation list to each student in a variety of ways, such as through the notification function of the online learning platform, email, or mobile application message reminders, so that students can easily view the recommended courses and choose the courses that are suitable for them to study in time.

[0077] The entire recommendation process takes into account the combination of students' individual characteristics and group characteristics. The course recommendation list finally pushed to students strikes a balance between personalization and collaboration, helping students better plan their music learning path and improve the consistency and effectiveness of learning.

[0078] On the other hand, the present invention also provides an online music course recommendation system based on data processing, comprising: A collection module, used for collecting first data of students, wherein the first data at least includes music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time; A processing module, configured to pre-process the first data to generate a multi-dimensional user portrait, and generate a user-course interaction matrix according to the multi-dimensional user portrait; A generating module, for determining the student's available time for music practice according to the course schedule and the historical music homework submission time; An aggregation module, for aggregating students into learning groups according to the music interest preference, music level, musical instrument ownership, and available time for music practice; A push module is used to generate a course recommendation list for each student based on the study group and the user-course interaction matrix, and push the course recommendation list to the student.

[0079] The prior art mentioned in the above background technology section and specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A method for recommending online music courses based on data processing, characterized in that: The method comprises the following steps: Collecting first data of the student, the first data at least including music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time; Preprocessing the first data to generate a multi-dimensional user portrait, and generating a user-course interaction matrix according to the multi-dimensional user portrait; Determine the student's available time for music practice based on the course schedule and the time for submission of historical music assignments; Aggregating students into study groups based on the musical interest preferences, musical level, musical instrument ownership, and available time for music practice; Based on the study group and the user-course interaction matrix, a course recommendation list is generated for each student, and the course recommendation list is pushed to the student.

2. The online music course recommendation method based on data processing according to claim 1 is characterized in that: The preprocessing includes: data cleaning, missing value filling, data standardization and normalization processing.

3. The online music course recommendation method based on data processing according to claim 1 is characterized in that: Generating a user-course interaction matrix according to the multi-dimensional user portrait includes: Combine the user's music interests, music level, and instrument ownership to generate a user feature vector ; in: The user's preference for music style; The user's music level; A musical instrument owned by the user; For each course, a feature vector is generated based on the content characteristics of the course. in: is the music style vector of the course, indicating the music styles and proportions involved in the course; Score the difficulty of the course, with a numerical value between 0 and 1, reflecting the musical level required by the course; is the vector of instrument types required for the course, indicating the types of instruments needed to learn the course; Calculate the user's music interest vector and the course music style vector : ， Calculate the user's music level score Difficulty rating of the course The degree of fit between: The match between the user's instrument ownership and the instrument required by the course is calculated using the vector dot product: ； Calculate the content-based rating of user 𝑢 for course 𝑐 : in, is the weight coefficient of different similarity scores, satisfying ; Furthermore, it captures the similarities between users and the commonality of their learning habits; Counting users With users Similarities between: ， in, is the rating of user 𝑖 on course 𝑗, is the average rating of user 𝑖, For users Average rating of Based on the historical ratings of similar users, About Courses Rating Make predictions: ， Generate the final user-course interaction matrix ; ， in, and is the fusion coefficient, satisfying .

4. The online music course recommendation method based on data processing according to claim 1 is characterized in that: Based on the course schedule and the submission time of historical music homework, the available time for students to practice music includes: Traverse all time periods of each day and mark all The time period is the student's free time period, and a list of all free time in a week is generated; Job submission frequency statistics: Statistics on the submission time of historical music homework, analyze the time pattern of students completing homework, and identify the students' habitual practice time periods; Use time series analysis to count the time points of homework submission, find out the time windows in a week when students frequently do music learning and homework, and generate a priority time period list; Determine the student's optimal practice time period A by finding the intersection of the free time in the course schedule and the priority time period of the homework ij , specifically expressed as: ， in: It means that the student’s 𝑗th time period on the 𝑖th day is an idle time period; represents the priority time for submitting a job in the 𝑗th time period of the 𝑖th day; Based on the list of optimal practice times generated by the intersection results, the time periods within a week are sorted according to the length of free time and the frequency indicators of priority time, and finally a timetable suitable for students to practice music is determined.

5. The online music course recommendation method based on data processing according to claim 1 is characterized in that: When aggregating students into study groups, the K-means clustering algorithm is used.

6. An online music course recommendation system based on data processing, characterized in that: The system includes the following modules: A collection module, used for collecting first data of students, wherein the first data at least includes music interest preference, music level, musical instrument ownership, course schedule, and historical music homework submission time; A processing module, configured to pre-process the first data to generate a multi-dimensional user portrait, and generate a user-course interaction matrix according to the multi-dimensional user portrait; A generating module, for determining the student's available time for music practice according to the course schedule and the historical music homework submission time; An aggregation module, for aggregating students into learning groups according to the music interest preference, music level, musical instrument ownership, and available time for music practice; A push module is used to generate a course recommendation list for each student based on the study group and the user-course interaction matrix, and push the course recommendation list to the student.

7. The online music course recommendation system based on data processing according to claim 6 is characterized in that: The preprocessing includes: data cleaning, missing value filling, data standardization and normalization processing.

8. The online music course recommendation system based on data processing according to claim 6 is characterized in that: Generating a user-course interaction matrix according to the multi-dimensional user portrait includes: Combine the user's music interests, music level, and instrument ownership to generate a user feature vector ; in: The user's preference for music style; The user's music level; A musical instrument owned by the user; For each course, a feature vector is generated based on the content characteristics of the course. in: is the music style vector of the course, indicating the music styles and proportions involved in the course; Score the difficulty of the course, with a numerical value between 0 and 1, reflecting the musical level required by the course; is the vector of instrument types required for the course, indicating the types of instruments needed to learn the course; Calculate the user's music interest vector With course music style vector : ， Calculate the user's music level score Difficulty rating of the course The degree of fit between: ， Calculate the user's musical instrument ownership The instruments required for the course The matching degree between them is calculated using the vector dot product: Calculate the content-based rating of user 𝑢 for course 𝑐 : in, is the weight coefficient of different similarity scores, satisfying ; Furthermore, it captures the similarities between users and the commonality of their learning habits; Counting users With users Similarities between: in, is the rating of user 𝑖 on course 𝑗, is the average rating of user 𝑖, For users Average rating of Based on the historical ratings of similar users, About Courses Rating Make predictions: Generate the final user-course interaction matrix in, and is the fusion coefficient, satisfying .

9. The online music course recommendation system based on data processing according to claim 6, characterized in that: Based on the course schedule and the submission time of historical music homework, the available time for students to practice music includes: Traverse all time periods of each day and mark all The time period is the student's free time period, and a list of all free time in a week is generated; Job submission frequency statistics: Statistics on the submission time of historical music homework, analyze the time pattern of students completing homework, and identify the students' habitual practice time periods; Use time series analysis to count the time points of homework submission, find out the time windows in a week when students frequently do music learning and homework, and generate a priority time period list; Determine the best time for students to practice by finding the intersection of the free time in the course schedule and the priority time of the homework A ij , specifically expressed as: in: It means that the student’s 𝑗th time period on the 𝑖th day is an idle time period; represents the priority time for submitting a job in the 𝑗th time period of the 𝑖th day; Based on the list of optimal practice times generated by the intersection results, the time periods within a week are sorted according to the length of free time and the frequency indicators of priority time, and finally a timetable suitable for students to practice music is determined.

10. The online music course recommendation system based on data processing according to claim 6, characterized in that: When aggregating students into study groups, the K-means clustering algorithm is used.

Images