Music teaching recommendation method and system based on knowledge graph

Abstract

The invention discloses a music teaching recommendation method and system based on a knowledge graph, and relates to the technical field of teaching recommendations, the method recommends a target node to a student according to each node music teaching resource in the knowledge graph and the learning record of the learned student in each node, and then according to the historical learning record of the student, the target node is recommended to the student. According to the method and the device, a music score recommendation mode of a student is confirmed, a target teaching music score is recommended to the student, and a learning period of the target teaching music score is confirmed according to a historical learning condition of the target teaching music score and learning conditions of the student in different periods, so that the adaptability of the music score and the learning period to the student is improved, and the music score recommendation accuracy and reliability are improved; personalized and targeted recommendation of the music score of the student is realized, the learning interest of the student is effectively stimulated, the learning requirements of different students are met, and the learning effect of the student is guaranteed, so that the music teaching recommendation effect is improved, and the satisfaction degree of the student is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of teaching recommendation, and in particular to a music teaching recommendation method and system based on knowledge graph. Background Art

[0002] The knowledge graph can integrate music scores of different difficulty levels. At the same time, it can recommend targeted music scores to students based on their interests and preferences, which can stimulate students' learning interest, meet the learning needs of different students, and ensure students' learning effects.

[0003] Prior art, such as a music teaching recommendation method and system based on knowledge graph disclosed in the application with publication number CN118245631A. The method comprises the following steps: obtaining students' music learning history data and performing course interest and hobby analysis and knowledge level assessment to obtain students' music knowledge level data; obtaining music teaching course information data and constructing a knowledge graph to obtain students' music course recommendation knowledge graph; performing music theory performance assessment and ability difficulty recommendation index calculation on music teaching course information data to obtain students' music course performance difficulty recommendation index; performing course recommendation processing on students' music course recommendation knowledge graph to obtain students' music teaching course recommendation result data; transmitting students' music teaching course recommendation result data to the cloud platform for personalized display to perform corresponding personalized music teaching course recommendation tasks. The present invention can achieve more accurate and personalized music course recommendations.

[0004] Prior art, such as a music teaching recommendation method and system based on knowledge graph disclosed in application publication number CN114780785A, the method includes: obtaining the historical teaching records of the target user, the historical teaching records include multiple historical music teaching content resources that the target user has obtained; obtaining the first knowledge graph, the first knowledge graph includes nodes corresponding to multiple music teaching content resources and node connections between each node; determining the target music teaching content resource; according to the first knowledge graph, determining the music teaching content resource to be recommended corresponding to the target music teaching content resource from the preset music teaching content resource library and pushing it to the target user. This application combines various feature data based on the characteristics of music teaching content resources, thereby improving the reliability and accuracy of music teaching content resource recommendation.

[0005] The above scheme has at least the following deficiencies: 1. The difficulty of learning the music scores in each node in the knowledge graph may be different. The above scheme only matches the teaching resources according to the students' situation, but the learning situation of all students who have studied in each node also reflects the adaptability of the teaching resources in each node to different types of students. The above scheme lacks analysis of the adaptability of the teaching resources of each node to different types of students based on the learning situation of other students, so it is impossible to accurately select suitable nodes for students, reducing the recommendation effect of teaching resources.

[0006] 2. When recommending teaching music scores for students, students' preferences for music types and authors affect their choice of music scores. The above scheme lacks analysis of students' preferences for music types and authors, and thus cannot select a music score recommendation scheme suitable for students, reducing the accuracy of music score recommendations for students, failing to effectively stimulate students' interest in learning, failing to meet the learning needs of different students, and failing to guarantee students' learning outcomes.

[0007] 3. Different music scores have different styles, and the learning effects in different time periods are also different. The above scheme lacks the selection of appropriate learning time periods based on different music scores and students' practice time preferences, thus failing to improve students' learning effects. Summary of the invention

[0008] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a music teaching recommendation method and system based on knowledge graph.

[0009] In order to solve the above technical problems, the present invention adopts the following technical solutions: The first aspect of the present invention provides a music teaching recommendation method based on knowledge graph, comprising the following steps: S1, obtaining basic data of students and constructing student portraits.

[0010] S2. Use the student's portrait to confirm the target node corresponding to the student from the music knowledge graph, and obtain the music teaching resources in the target node.

[0011] S3. Obtain the student's historical learning record, and then select the target teaching music score corresponding to the student from the music teaching resources of the target node.

[0012] S4. Send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, S3 is executed again. If the student chooses to confirm, a teaching plan for the student is formulated according to the target teaching music score corresponding to the student.

[0013] The second aspect of the present invention provides a music teaching recommendation system based on knowledge graph, including: a student acquisition module, used to obtain basic data of students and construct student portraits.

[0014] The node matching module is used to use the student's portrait to confirm the target node corresponding to the student from the music knowledge graph and obtain the music teaching resources in the target node.

[0015] The music score recommendation module is used to obtain students' historical learning records and then select the target teaching music score corresponding to the students from the music teaching resources of the target node.

[0016] The teaching formulation module is used to send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, the music score will be recommended again. If the student chooses to confirm, the student's teaching plan will be formulated according to the student's corresponding target teaching music score.

[0017] The beneficial effects of the present invention are: 1. The present invention provides a music teaching recommendation method and system based on knowledge graph, which recommends target nodes to students according to the music teaching resources of each node in the knowledge graph and the learning records of the students who have learned in each node, and then confirms the method of recommending students' music scores according to the students' historical learning records, recommends target teaching music scores to students, and then confirms the learning period of the target teaching music scores according to the historical learning situation of the target teaching music scores and the learning situation of the students in different time periods, improves the adaptability of the music scores and learning period to the students, improves the accuracy and reliability of music score recommendations, realizes personalized and targeted recommendations for students' music scores, effectively stimulates students' learning interest, meets the learning needs of different students, and ensures students' learning effects, thereby improving the effect of music teaching recommendations and improving students' satisfaction.

[0018] 2. This application recommends target nodes for students based on age groups and professional ability levels, improves the adaptability of the age groups and professional ability levels suitable for the teaching resources in the nodes to the age and professional ability levels of the students, improves the matching of the node teaching resources with the students, and makes the teaching resources allocated to the students more suitable for the students.

[0019] 3. This application confirms the method of recommending student music scores based on the student's historical learning records and recommends target teaching music scores to students. It can ensure the accuracy of the student music score recommendations, realize personalized and diverse music score recommendations, ensure students' satisfaction with the music scores, and improve students' learning interest.

[0020] 4. According to the historical learning situation of the target teaching music score and the students' learning situation in different periods of time, confirm the learning period of the target teaching music score, and choose a period with better learning effect for students to study, improve the learning effect, make it easier for students to have an optimistic learning attitude, and increase the attractiveness of music score learning to students. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] 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 or the description of the prior art 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.

[0022] Figure 1 The present invention is a schematic flow chart of the steps for implementing the method.

[0023] Figure 2 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Embodiment 1: See also Figure 1 As shown, a music teaching recommendation method based on knowledge graph includes the following steps: S1. Obtaining basic data of students and constructing student portraits.

[0026] Among them, when students register and log in to the music teaching recommendation system for the first time, they are prompted to fill in basic data, including name, age, and professional ability level.

[0027] The student portrait includes age group and professional ability level.

[0028] S2. Use the student's portrait to confirm the target node corresponding to the student from the music knowledge graph, and obtain the music teaching resources in the target node.

[0029] In the above, the staff of the music teaching recommendation system collected multiple music scores, set a professional ability level for each music score, and recorded a teaching video for each music score. Then, all the music scores and teaching videos with the same professional ability level were regarded as music teaching resources in one node, and the professional ability level of all the music scores in each node was used as the recommended professional ability level. At the same time, each node was connected in sequence according to the recommended professional ability level to form a music knowledge graph.

[0030] In a specific embodiment, the target node corresponding to the student is confirmed from the music knowledge graph, and the specific process is as follows: obtaining the recommended professional ability level corresponding to each node in the music knowledge graph and the professional ability level and age group of each student in each node, and at the same time obtaining the evaluation data of each student in the communication area of ​​each node, and analyzing the recommended weight of each professional ability level and the recommended weight of each age group in each node.

[0031] Using the age group and professional ability level in the student portrait, we can obtain the recommended weight of the student's professional ability level and age group in each node, which are recorded as and , where x represents the number of each node and x is a positive integer.

[0032] Preferably, the recommendation weight of each professional ability level and the recommendation weight of each age group in each node are analyzed, and the specific process is as follows: according to the professional ability level of each student in each node, the students of each professional ability level in each node are screened out, and then the historical learning records of each student of each professional ability level in each node are obtained from the background learning records, and then the practice time and learning performance of each music score are obtained from the learning records.

[0033] It should be noted that after students have finished learning the teaching video of the music score, they will practice freely. After the free practice, the students will record and upload the music score performance video. The music teaching recommendation system will send the music score performance video to the teacher who recorded the music score teaching video. The teacher will score the music score performance video, and the score will be the learning performance. At the same time, the time between the time when the student starts learning the music score teaching video and the time when the music score performance video is uploaded is used as the practice time. Then the music teaching recommendation system records the student's practice time and learning performance when learning the music score.

[0034] After removing the maximum and minimum values ​​of the practice time and learning performance of each student of each professional ability level in each node corresponding to each score, the average practice time and average learning performance of the score corresponding to each professional ability level in each node are obtained, and then the student learning index of each professional ability level in each node is calculated, which is recorded as , y represents the number of each professional ability level, and y is a positive integer.

[0035] In the above, , where , They represent the average practice time and average learning performance of the music score corresponding to the yth professional ability level in the xth node, , They respectively represent the average practice time and average learning performance of the music scores in the x-th node.

[0036] in , The average practice time and average learning performance of the music scores corresponding to the yth professional ability level in the xth node are calculated by the mean.

[0037] Obtain the evaluation data of each student of each professional ability level in each node, obtain the evaluation content and recommendation score from the evaluation data, and calculate the student recommendation index of each professional ability level in each node, which is recorded as .

[0038] It should be noted that a student exchange area is set up in each node, and students can score and evaluate the music scores they have learned in the student exchange area. After students score multiple music scores in a node, the average value of the multiple music scores is the student's recommended score for the node. All evaluations of students in each node are integrated, and then the word segmentation technology is used to obtain the number of words in each node, and then the part-of-speech tagging is used to obtain the number of positive and negative words in each node.

[0039] The recommended scores, word counts, positive word counts, and negative word counts of each student at each professional ability level in each node are counted, and then the average is calculated to obtain the recommended scores, word counts, positive word counts, and negative word counts of students at each professional ability level in each node, which are recorded as , , and ; Substitute into the analysis formula , get the student recommendation index of the yth professional ability level in the xth node , where TJ is the passing score set, , They are respectively the threshold of the positive word ratio and the threshold of the negative word ratio.

[0040] Among them, the passing score is 60% of the maximum score. Assuming that the recommended score range is 0-100, the passing score is 60. The threshold of the proportion of positive words and the threshold of the proportion of negative words are set by the staff.

[0041] Using the analytical formula: , get the recommended weight of the yth professional ability level in the xth node , where s represents the number of professional ability levels.

[0042] The recommended weights for each age group in each node are calculated in the same way as the recommended weights for each professional ability level.

[0043] Using matching analysis formula expression: , get the matching coefficient between the student and the xth node , where ZY represents the professional ability level of students, represents the recommended professional competence level of the xth node, is the set professional ability level error value, and e represents a natural constant.

[0044] Among them, the professional ability level error value is set by the staff, indicating the difference between the recommended professional ability level and the permitted professional ability level.

[0045] S3, obtaining the student's historical learning record, and then selecting the target teaching music score corresponding to the student from the music teaching resources of the target node; It should be noted that the students' historical learning records are obtained from the background of the music teaching recommendation system.

[0046] In a specific embodiment, the target teaching music score corresponding to the student is selected from the music teaching resources of the target node, and the specific process is as follows: the student's learned music scores, the music type and the author of each learned music score are obtained from the student's historical learning records.

[0047] The music types of the students' music scores are used to analyze the correlation between the music types of the students' music scores, where the correlation includes values ​​of 1 and 0. When the correlation is 1, it indicates that the music types of the students' music scores are strongly correlated, otherwise it indicates that the music types of the students' music scores are weakly correlated.

[0048] In the above, the correlation analysis process of the music type of the students' music scores is as follows: the music elements of each learned music score are extracted by using audio processing technology, and according to the music type of each learned music score, the music elements corresponding to each music type in the students' learned music scores are counted, and then the music elements corresponding to each music type are used to calculate the similarity between each music type using cosine similarity, and then the mean is calculated to obtain the average similarity of the music type of the students' music scores, denoted as A1, and the similarity threshold is set at the same time, denoted as A, then the correlation expression is: , where Indicates the relevance of the student's music score to the type of music.

[0049] It should be noted that the similarity threshold is the critical value for judging whether the music types are related, which is set by the staff. For example, if the similarity threshold is 5 and the average similarity of the music types of the students' scores is 7, 7>5, it indicates that the music types of the students' scores are highly related.

[0050] Among them, music types include classical music, Latin music and pop music, and music elements include melody, rhythm, harmony and timbre.

[0051] Cosine similarity evaluates the similarity of two vectors by calculating the cosine value of the angle between them. The specific analysis method has been made public and will not be repeated here.

[0052] The correlation between the authors of the students' music scores is analyzed by using the authors of the music scores that the students have learned. The correlation includes the values ​​of 1 and 0. When the correlation is 1, it indicates that the correlation between the authors of the students' music scores is strong, otherwise it indicates that the correlation between the authors of the students' music scores is weak.

[0053] Preferably, the analysis process of the correlation between the authors of students' music scores is as follows: based on the Internet, the correlation data of the authors of all music scores in the knowledge graph is obtained, and then the correlation data of the authors of all music scores are modeled and integrated using semantic web technology to construct an author relationship graph, in which one node represents one author.

[0054] According to the authors of the music scores that the students have learned, the authors that the students have learned are obtained. According to the author relationship map, the nodes corresponding to the authors that the students have learned are obtained, and then the number of interval nodes between the nodes corresponding to the authors is obtained. Then, the number of interval nodes between the nodes corresponding to the authors is averaged, and the average number of interval nodes between the authors is obtained, which is recorded as M1. A threshold value for the number of interval nodes is set, which is recorded as M. Using the expression: , get the relevance of the student's music score author.

[0055] The process of setting the interval node quantity threshold and the similarity threshold is the same and will not be described again.

[0056] Semantic Web technology can understand words and concepts, and also understand the logical relationships between them, making communication more efficient and valuable. The specific implementation process has been made public and will not be repeated here.

[0057] The author's associated data includes the author's date of birth and place of residence, etc.

[0058] When the correlation between the student's music score type is 0 and the correlation between the music score author is 0, a one-category music score selection scheme is used; when the correlation between the student's music score type is 1 and the correlation between the music score author is 0, a two-category music score selection scheme is used; when the correlation between the student's music score type is 0 and the correlation between the music score author is 1, a three-category music score selection scheme is used; when the correlation between the student's music score type is 1 and the correlation between the music score author is 1, a four-category music score selection scheme is used, and then the target teaching music score corresponding to the student is selected from the music teaching resources of the target node.

[0059] In the above, one type of music score selection scheme is: remove the music types of the music scores that the students have learned from the music types of the music scores in the target node, and at the same time remove the authors of the music scores that the students have learned from the authors of the music scores in the target node to obtain the remaining music scores.

[0060] Obtain the historical student academic records of the music types that students have learned, analyze the students' preference weights for each label in the music type of the music score, and also analyze the students' preference weights for each label in the music author.

[0061] It should be noted that the historical student academic records of the music types that the students have learned are obtained from the background of the music teaching recommendation system.

[0062] Among them, the analysis process of students' preference weights for each label in the music type of the score is as follows: each label in the music type includes niche music type, ordinary music type and popular music type; the number of students who have learned each music type is obtained from the historical student academic records of the music type of each music type that the students have learned, and when the number of students who have learned a certain music type is less than the preset lower limit of the number of students who have learned, the label of the music type is the niche music type; when the number of students who have learned a certain music type is greater than the preset lower limit of the number of students who have learned, and less than the preset upper limit of the number of students who have learned, the label of the music type is the ordinary music type; the number of students who have learned a certain music type is greater than the preset upper limit of the number of students who have learned, then the label of the music type is the popular music type; in this way, the labels corresponding to each music type are obtained, and the labels of the music types corresponding to each music score that the students have learned are obtained in advance, and the number of scores of niche music type, ordinary music type and popular music type in the music type that the students have learned is counted, and then the number of scores of niche music type, ordinary music type and popular music type are divided by the total number of scores that have been learned, respectively, to obtain the preference weights of niche music type, ordinary music type and popular music type.

[0063] It should be noted that the setting process of the upper limit and lower limit of the number of learned students and the similarity threshold is the same and will not be repeated here.

[0064] In the above, the analysis process of students' preference weights for each label in the music score author is as follows: the author labels include niche authors, ordinary authors and popular authors; the number of students who have learned the scores of the authors corresponding to each score they have learned is obtained from the background of the music teaching recommendation system, and then according to the analysis process of students' preference weights for each label in the music score type, the students' preference weights for each label in the music score author are analyzed.

[0065] Get the music type labels and author labels corresponding to each remaining score, then extract the students' preference weights for the music type labels and author labels corresponding to each remaining score, and then add the students' preference weights for the music type labels and author labels corresponding to each remaining score to get the students' preference weights for each remaining score, and select the remaining score corresponding to the maximum preference weight as the target teaching score.

[0066] Preferably, the second category of music score selection scheme is: obtain a music type association table from a database, obtain the students' learned music types according to the music types of the students' learned music scores, use the music type association table to obtain the music types whose association with the learned music types is greater than a preset association value as the optional music types, obtain the music type of each music score from the music teaching resources of the target node, and compare them with the optional music types, and select the music scores with the same music types as the optional music types as the optional music types as the optional music scores.

[0067] It should be noted that the setting process of the music type association table is: use audio processing technology to extract each music element in each graph of each music type in the knowledge graph, use cosine similarity to calculate the similarity between each music type for each music element in each graph of each music type, and use it as the association between each music type, and then construct a music type association table.

[0068] Using the music type association table, obtain the association between the music type of each optional score and the music type of each learned score, then add them up to get the total association between the music type of each optional score and the music type of the learned score, and select the optional score with the largest total association as the target teaching score.

[0069] Preferably, the three types of music score selection schemes are as follows: obtaining the author of each music score from the music teaching resources of the target node, and using the authors of the music scores that the students have learned, counting the number of music scores learned by the students from each author, and calculating the priority coefficient of the students learning from each author; comparing the author of each music score with the authors learned by the students, if the author of a music score is the same as an author learned by the student, then using the music score as the preliminary music score, thereby obtaining the preliminary music scores.

[0070] Sort the preliminary music scores according to the priority coefficients of the authors of the preliminary music scores that the students learn. If there are at least two preliminary music scores with the same ranking, then sort the preliminary music scores with the same ranking according to the number of times they have been learned, so as to obtain the ranking of the preliminary music scores, and take the preliminary music score with the first ranking as the target teaching music score.

[0071] Preferably, the four-category music score selection scheme is a combination of the two-category music score selection scheme and the three-category music score selection scheme. The two-category music score selection scheme and the three-category music score selection scheme have been explained above and will not be repeated here.

[0072] S4. Send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, S3 is executed again. If the student chooses to confirm, a teaching plan for the student is formulated according to the target teaching music score corresponding to the student.

[0073] In a specific embodiment, the specific process of formulating a teaching plan for students is as follows: the learning period and learning performance of each learned music score are obtained from the student's corresponding historical learning records, and the learning period, learning performance, age group and professional ability level of each learned student in the target teaching music score are obtained, and each learned student whose age group and professional ability level correspond to the student's age group and professional ability level is selected as each reference student.

[0074] Using the learning periods and learning results of each reference student, analyze the learning effect coefficient of each learning period, and select the learning periods with learning effect coefficients greater than the preset learning effect coefficient threshold as the optional periods; at the same time, use the learning periods and learning results of each student's learned music score to analyze the students' learning effect coefficients in each learning period, and select the learning periods with learning effect coefficients greater than the preset learning effect coefficient threshold as the marked periods.

[0075] In the above, according to the study period and study results of each reference student, the study results corresponding to each study period are counted and recorded as , t represents the number of each learning period, r represents the number of each learning score, t and r are both positive integers; Substitute into the analysis formula , get the learning effect coefficient of the tth learning period , where p represents the number of learning scores, and Q represents the set passing score.

[0076] The learning period can be 9:00-11:00, 14:00-16:00, 16:00-18:00, etc. The setting process of the learning effect coefficient threshold is the same as the setting process of the similarity threshold, which will not be repeated here. The setting process of the passing grade is the same as the passing score, which will not be repeated here.

[0077] Get the time difference between each optional time period and each marked time period, calculate the difference rate between each optional time period and each marked time period, when there is at least one difference rate less than the difference rate threshold, select the marked time period with the smallest difference rate as the target learning time period, when there is no difference rate less than the difference rate threshold, select the optional time period corresponding to the maximum learning effect coefficient as the target learning time period, and the teaching plan is composed of the target teaching score and the target learning time period.

[0078] In the above, a time difference threshold is set, and the time difference between each optional time period and each marked time period is subtracted from the time difference threshold and then divided by the time difference threshold to obtain the difference rate between each optional time period and each marked time period.

[0079] The setting process of the time difference threshold and the difference rate threshold is the same as that of the similarity threshold, which will not be described in detail here.

[0080] Embodiment 2: See also Figure 2As shown, a music teaching recommendation system based on knowledge graph includes: a student acquisition module, a node matching module, a music score recommendation module, a teaching formulation module and a database.

[0081] The student acquisition module is used to obtain students' basic data and build student portraits.

[0082] The node matching module is used to use the student's portrait to confirm the target node corresponding to the student from the music knowledge graph and obtain the music teaching resources in the target node.

[0083] The music score recommendation module is used to obtain students' historical learning records and then select the target teaching music score corresponding to the students from the music teaching resources of the target node.

[0084] The teaching formulation module is used to send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, the music score will be recommended again. If the student chooses to confirm, the student's teaching plan will be formulated according to the student's corresponding target teaching music score.

[0085] The database is used to store the music type association table.

[0086] The embodiment of the present invention recommends target nodes to students based on the music teaching resources of each node in the knowledge graph and the learning records of the students who have learned in each node, and then confirms the method of recommending students' music scores based on the students' historical learning records, recommends target teaching music scores to students, and then confirms the learning period of the target teaching music scores based on the historical learning situation of the target teaching music scores and the learning situation of the students in different time periods, thereby improving the adaptability of the music scores and learning period to the students, improving the accuracy and reliability of music score recommendations, and realizing personalized and targeted recommendations for students' music scores, effectively stimulating students' learning interest, meeting the learning needs of different students, and ensuring students' learning effects, thereby improving the effect of music teaching recommendations and improving students' satisfaction.

[0087] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A music teaching recommendation method based on knowledge graph, characterized in that: The steps include: S1. Obtain students’ basic data and build student portraits; S2. Using the student's portrait, identify the target node corresponding to the student from the music knowledge graph, and obtain the music teaching resources in the target node; S3, obtaining the student's historical learning record, and then selecting the target teaching music score corresponding to the student from the music teaching resources of the target node; The specific process of confirming the target node corresponding to the student from the music knowledge graph is as follows: Obtain the recommended professional ability level corresponding to each node in the music knowledge graph, as well as the professional ability level and age group of each student in each node, and at the same time obtain the evaluation data of each student in the communication area of ​​each node, and analyze the recommended weight of each professional ability level and the recommended weight of each age group in each node; Using the age group and professional ability level in the student portrait, we can obtain the recommended weight of the student's professional ability level and age group in each node, which are recorded as and , where x represents the number of each node and x is a positive integer; Using matching analysis formula expression: , get the matching coefficient between the student and the xth node , where ZY represents the professional ability level of students, represents the recommended professional competence level of the xth node, is the error value of the set professional ability level, and e represents a natural constant; The specific process of analyzing the recommended weights of each professional ability level and each age group in each node is as follows: According to the professional ability level of each student in each node, each student of each professional ability level in each node is screened out, and then the historical learning records of each student of each professional ability level in each node are obtained from the background learning records, and then the practice time and learning results of each score are obtained from the learning records; After removing the maximum and minimum values ​​of the practice time and learning performance of each student of each professional ability level in each node corresponding to each score, the average practice time and average learning performance of the score corresponding to each professional ability level in each node are obtained, and then the student learning index of each professional ability level in each node is calculated, which is recorded as , y represents the number of each professional ability level, and y is a positive integer; Obtain the evaluation data of each student of each professional ability level in each node, obtain the evaluation content and recommendation score from the evaluation data, and calculate the student recommendation index of each professional ability level in each node, which is recorded as ; Using the analytical formula: , get the recommended weight of the yth professional ability level in the xth node , where s represents the number of professional ability levels; The recommended weights for each age group in each node are calculated in the same way as the recommended weights for each professional ability level; The specific process of selecting the target teaching music score corresponding to the student from the music teaching resources of the target node is as follows: Acquire each music score that the student has learned, the music type and the author of each music score that the student has learned from the student's historical learning record; Using the music types of the music scores that the students have learned, the correlation of the music types of the music scores of the students is analyzed, wherein the correlation includes the values ​​of 1 and 0. When the correlation is 1, it indicates that the music types of the music scores of the students are highly correlated, and vice versa, it indicates that the music types of the music scores of the students are relatively weakly correlated; Using the authors of the music scores that the students have learned, the correlation of the authors of the students' music scores is analyzed, where the correlation includes the values ​​of 1 and 0. When the correlation is 1, it indicates that the correlation of the authors of the students' music scores is strong, otherwise it indicates that the correlation of the authors of the students' music scores is weak; When the correlation of the music type of the student's music score is 0 and the correlation of the music author is 0, a one-category music score selection scheme is used; when the correlation of the music type of the student's music score is 1 and the correlation of the music author is 0, a two-category music score selection scheme is used; when the correlation of the music type of the student's music score is 0 and the correlation of the music author is 1, a three-category music score selection scheme is used; when the correlation of the music type of the student's music score is 1 and the correlation of the music author is 1, a four-category music score selection scheme is used, and then the target teaching music score corresponding to the student is selected from the music teaching resources of the target node; The first music score selection scheme is as follows: the music types of the music scores in the target node are removed from the music types of the music scores that the students have learned, and the authors of the music scores in the target node are removed from the authors of the music scores that the students have learned, to obtain the remaining music scores; Obtain the historical student academic records of the music types that students have learned, analyze the students' preference weights for each label in the music type of the music score, and analyze the students' preference weights for each label in the author of the music score; Obtain the music type labels and author labels corresponding to the remaining scores, then extract the students' preference weights for the music type labels and author labels corresponding to the remaining scores, and then add the students' preference weights for the music type labels and author labels corresponding to the remaining scores to obtain the students' preference weights for the remaining scores, and select the remaining scores corresponding to the maximum preference weights as the target teaching scores; The second type of music score selection scheme is as follows: obtaining a music type association table from a database, obtaining each learned music type of the student according to the music type of each learned music score of the student, using the music type association table, obtaining each music type with a correlation degree greater than a preset correlation value with each learned music type as each optional music type, obtaining the music type of each music score from the music teaching resources of the target node, and comparing it with each optional music type, and selecting each music score with the same music type as each optional music type as each optional music type as each optional music score; Using the music type association table, obtain the association between the music type of each optional music score and the music type of each learned music score, then add them up to obtain the total association between the music type of each optional music score and the music type of the learned music score, and select the optional music score with the largest total association as the target teaching music score; The three types of music score selection schemes are as follows: obtaining the author of each music score from the music teaching resources of the target node, and using the authors of the music scores that the students have learned, counting the number of music scores that the students have learned from each author, and calculating the priority coefficient of the students learning from each author; comparing the author of each music score with the authors that the students have learned, if the author of a music score is the same as the author that the student has learned, then the music score is used as the preliminary music score, thereby obtaining each preliminary music score; Sort the preliminary selected scores according to the priority coefficients of the authors of the preliminary selected scores that the students learn, where if there are at least two preliminary selected scores with the same ranking, then sort the preliminary selected scores with the same ranking according to the number of times they have been learned, so as to obtain the ranking of the preliminary selected scores, and use the first ranked preliminary selected score as the target teaching score; S4. Send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, S3 is executed again. If the student chooses to confirm, a teaching plan for the student is formulated according to the target teaching music score corresponding to the student.

2. A music teaching recommendation method based on knowledge graph according to claim 1, characterized in that: The student portrait includes age group and professional ability level.

3. According to the music teaching recommendation method based on knowledge graph in claim 1, it is characterized in that: The specific process of formulating a teaching plan for students is as follows: Obtain the learning period and learning results of each learned score from the student's corresponding historical learning records, and at the same time obtain the learning period, learning results, age group and professional ability level of each learned student in the target teaching score, and select each learned student whose age group and professional ability level correspond to the student's age group and professional ability level as each reference student; Using the learning period and learning results of each reference student, analyzing the learning effect coefficient of each learning period, selecting the learning period with a learning effect coefficient greater than a preset learning effect coefficient threshold as each optional period; at the same time, using the learning period and learning results of each learned music score of the student, analyzing the learning effect coefficient of the student in each learning period, and selecting the learning period with a learning effect coefficient greater than the preset learning effect coefficient threshold as each marked period; Get the time difference between each optional time period and each marked time period, calculate the difference rate between each optional time period and each marked time period, when there is at least one difference rate less than the difference rate threshold, select the marked time period with the smallest difference rate as the target learning time period, when there is no difference rate less than the difference rate threshold, select the optional time period corresponding to the maximum learning effect coefficient as the target learning time period, and the teaching plan is composed of the target teaching score and the target learning time period.

4. A music teaching recommendation system that implements the music teaching recommendation method based on knowledge graph according to any one of claims 1 to 3, characterized in that: include: The student acquisition module is used to obtain students' basic data and build student portraits; The node matching module is used to use the student's portrait to identify the target node corresponding to the student from the music knowledge graph and obtain the music teaching resources in the target node; The music score recommendation module is used to obtain the student's historical learning records and then select the target teaching music score corresponding to the student from the music teaching resources of the target node; The teaching formulation module is used to send the target teaching music score corresponding to the student to the student's user terminal for confirmation. If the student chooses to change, the music score will be recommended again. If the student chooses to confirm, the student's teaching plan will be formulated according to the student's corresponding target teaching music score.

Images