Human-Centered Computational Intelligence Information SystemsView this Special Issue
A Music Teaching Resource Management Model Based on Fuzzy Clustering Algorithm
This study develops and implements a music instructional resource platform based on fuzzy clustering algorithm knowledge in order to make the management of music teaching quality more scientific and standardised. This study presents the construction scheme of a music instructional resource management system in a higher education institution through requirements analysis, function design, system development and implementation, and system testing of the software system. In this study, a system is developed that uses computational intelligence to calculate the content and topics contained in resources based on semantic similarity, generate category labels, and cluster educational resources in order to integrate educational resources and implement effective management. It not only meets the needs of users who want to access the platform system via mobile terminals but also increases the platform system’s utilisation rate. Experiments show that this system’s stability can reach 95.37%. The retrieval accuracy is 96.24%, which is 8.94% better than the traditional method. It has the ability to effectively provide high-quality music teaching resources. Using the instructional resource system proposed in this study will improve teaching effectiveness, reduce learners’ learning burden, and increase learning efficiency.
Education has progressed rapidly in tandem with the country’s economic and cultural development. Schools and society have paid more attention to music courses that are popular with students . Art education is an important way for schools to provide quality education. On the one hand, modern society’s information construction integrates previously dispersed resources so that some can be reused and publicly used, greatly improving resource utilisation; on the other hand, it improves communication efficiency between people and lowers communication costs, greatly improving production efficiency. Information-based education considers information to be a fundamental component of the educational system, and it employs modern information technology in the field of education in a comprehensive and deep way to promote educational reform and development . Information education will inevitably result in the formation of a brand-new educational form: information education. Education informatization is a critical component of educational modernization, and all countries place a high value on its advancement. The construction of educational resources is at the heart of the modernization and informatization of educational content . It builds an efficient and convenient learning platform for learners by combining advanced computer technology, communication technology, and network technology. The development and integration of music teaching materials resources have become a critical way for schools to improve music teaching in the new era. At the same time, music curriculum resources come in a wide variety of formats, and the vast Internet platform offers a wealth of materials for integrating music curriculum resources . The abundance of music information and resources on the Internet has made it easier for teachers and students to learn more about music, but they still lament the lack of information resources specifically related to music major classroom teaching. Although the rapid growth of online instructional resources provides users with more options, the phenomenon of “resource overload” and “resource loss” raises the cost of finding the resources they need.
Educational resources play an important role in the classroom. Excellent educational resources can enhance teaching effectiveness, reduce learners’ learning burden, and increase learning efficiency . It can be developed and utilised, and it can also serve music teaching, whether it is explicit or implicit online music resources that can stimulate students’ music learning, cultivate students’ good musical aesthetic concepts and artistic sentiments, and truly promote the harmonious development of students’ body and mind. People paid too much attention to the application of technology in the past when building educational resources; that is, more emphasis was placed on the advancement of technology and less consideration was given to the universality of society and lack of humanistic care. This has had a significant impact on the effectiveness of educational resources. When using network information resources in music teaching, many schools are still in the early stages of searching for music materials and simply processing them online. To carry out information construction, most institutions of higher learning still use separate departments or business systems as units; each department has not formed a unified road map for construction planning. As a result, after each business system is built, multiple system architectures may exist, with their respective data structures and interface standards being completely inconsistent; as a result, each business system cannot be interconnected, forming an information island. As a result, realizing the sharing, knowledge, and visit of mobile terminal users to the platform of instructional resources in institutions of higher learning, effectively integrating instructional resources, and improving the efficiency of users’ use of various services in the system are of great scientific theoretical value and practical significance. Fuzzy clustering [6, 7] is a fuzzy mathematics method for classifying target objects using similarity relations. The fuzzy clustering algorithm is used to build and apply the music instructional resource platform in this study. The following are the article’s innovations:(1)This study puts forward an implementation scheme of music instructional resources construction platform based on a fuzzy clustering algorithm and discusses the core technologies in the implementation process in detail. In this study, the semantic grid is introduced to solve the problems of high-feature dimension and semantic connection between features caused by the clustering of online educational resources, so as to improve the accuracy of music educational resources retrieval and promote the effective utilisation of music educational resources.(2)In this study, the Mahout tool is used to realize the clustering analysis of resource data and establish the data model, and the personalized service mechanism is adopted in the system. On the one hand, the personalized initial resource selection of new users of the information platform is effectively realized; on the other hand, it effectively realizes the personalized update resource selection of the old users of the information platform. Experiments show that this system can better meet the user’s experience.
This study is divided into five sections based on the actual requirements. The introduction is the first section. This section provides an overview of the topic’s background, study structure, and research methods. The literature review is the second section. This section summarises relevant literature from both domestic and international sources, as well as the study’s research ideas and methods. The method section is the third part. The concept, types, and characteristics of music education resources are discussed in this section. The current state of music education resource development is analyzed; based on the deficiencies, some strategies and suggestions for the development and integration of a music education resource database were proposed. The model for music educational resources is built. The fourth section conducts an experiment with the model developed in this study, analyzing its performance and practical application impact. The fifth section is the conclusion, which summarises the entire text while also looking ahead to the future.
2. Related Work
Education does not refer only to manpower, material resources, and financial resources. With the appearance and development of distance online education and online education, the extension scope of educational resources has become wider. The construction of instructional resources has become one of the key research topics of educators. At present, the academic circles have done some research on the construction of music education resources.
Jennifer proposed to combine the advantages of traditional teaching mode and constructivist teaching mode, not only play the leading role of teachers to guide, inspire, and monitor the teaching process, but also fully reflect the initiative, enthusiasm, and creativity of students as the main body of the learning process. It further expounds on the advantages and application of streaming media technology in the construction of music instructional resources . Zandén and Thorgersen proposed a recommendation method based on fuzzy clustering, combined with collaborative filtering, intelligent word segmentation, and mobile agent technology, to predict and recommend instructional resources for target users’ retrieval expectations . The experimental results show that its recommendation quality and accuracy have been greatly improved. East and Hutchinson improved the traditional personalized information service technology by using clustering and proposed a feasible specific implementation scheme of a personalized recommendation system for educational resources based on user clustering . Andrews and Aubyn analyzed the user behavior of the instructional resource database based on the archives instructional resource database . Its research results show that the usefulness and ease of use of the system have the greatest impact on users’ willingness to use it; that is, the quality of instructional resources directly determines the user experience. Grunspan et al. took the music instructional resource management system as the starting point and conducted in-depth discussions and research on the informatization construction of institution of higher learning, especially the data management of instructional resources . Arnaiz-González et al. proposed a user interest representation method based on context and temporal vector space model and a resource description representation method based on context and cognitive level vector space model. The system adopts decision trees, fuzzy matching and reasoning, and ant colony clustering as supporting technologies for personalized services . Kim proposed a standard instructional resource library model in view of the low degree of resource abstraction and poor correlation between resources in most current instructional resource libraries. Through this model, a metadata index is established for each instructional resource object; the related educational resource objects are automatically associated through the metadata, and a semantic-based search function is provided . Gibson uses a subspace algorithm for cluster analysis . Its advantages are that it can automatically discover data with higher dimensions, and it has good and good scalability. Tucker et al. expounded on the current situation of the construction of online music education resources, analyzed the ways of using online music education resources, and looked forward to the prospect of using online music education resources to serve music teaching . Yuan proposed a network instructional resource sharing system based on the Hadoop platform . The system stores instructional resources on HDFS, which can be easily expanded. When the amount of data increases to a certain extent, the big data processing tools of Had00p can be used for data mining. Braun pointed out that only by strengthening the awareness of online music resources, making a reasonable application in music teaching practice, promoting the construction of curriculum teaching, and improving the efficiency of music teaching, the development of music teaching in the entire education system can be promoted . Gupta proposed a model to encourage users to participate in the creation of instructional resources based on self-organisation theory . This model is consistent with the Web 2.0 idea that data is created, utilised, and shared by users. A system in which users are forced to accept content is difficult to satisfy with systems such as educational resource libraries.
The related literature are thoroughly examined in this study, and clustering-based personalized information service technology is discussed. On the basis of the foregoing, a fuzzy clustering algorithm-based music instructional resource management platform is developed. It primarily makes use of the Python programming language, My SQL for data storage, and a Web crawler to find online educational resources. At the same time, the fuzzy clustering algorithm is used to cluster and analyze the data of music instructional resources, resulting in a more perfect feasibility for the use of music instructional resources in higher education institutions. In addition, the task scheduling in the scheduling server is optimised, which effectively solves the resource allocation problem between resource users and distributed distribution centres, resulting in system load balance. The test results indicate that the music instructional resource management platform system described in this study is stable and expandable. It also has a robust set of features and a well-defined structure, making it a viable scheme for managing future educational resources.
3.1. Music Education Resources
With the rapid development of computer and information technology, people have entered the information age. The six elements of the information system are as follows: information resources, information network, information technology application, information technology and industry, information technology talents, and information technology policies, regulations, and standards. The rapid development of computer multimedia technology and network technology has promoted the process of educational informatization. Followed by the vigorous construction of instructional resources, the number of online instructional resources has become extremely large. For an educational informatization construction, the information network is the foundation, and the instructional resources are the core. The utilisation of instructional resources and the application of information technology are the goals, and the information technology talents, information technology industry, and information technology policies, regulations, and standards are the guarantee. Personalized information service is a service based on modern humanities and information technology. With the help of advanced information technologies such as intelligent technology [20, 21], integrated technology, expert system technology, and information mining technology, knowledge products must be tailored for each user according to the diverse information needs of users. The information platform finally achieves the goal of connecting everything, that is, connecting people with people, connecting people with things, connecting things with things, and realizing a leap in production efficiency. Information-based education has the remarkable characteristics of multimedia teaching materials, resource globalization, teaching individualization, learning autonomy, activity cooperation, management automation, and environmental virtualization, among which educational resources are the core of all applications. Actively use the advantages of music resources to reasonably integrate and configure music instructional resources. The ultimate goal is to enrich the school’s music classrooms, enrich students’ artistic life, strengthen students’ music practice ability, improve students’ creative thinking, and cultivate students of self-cultivation and sentiment. Educational resources, as the core of educational informatization construction, play an important role in improving the quality of education and teaching and tapping the potential of educational development. Educational resources refer to all kinds of information resources that contain specific educational information, which can be provided to learners and can help and promote their learning, especially the instructional resources that can be transmitted on the Internet by digital signals. Digitization and educational value are its core features.
Music education resources that rely on paper books, newspapers, publications, and other physical carriers, as well as analogue radio carriers such as radio and television, are considered traditional. The network music education resource is a new type of music education resource that is preserved and disseminated through virtual digital technology and the Internet that it has created. Educational resources, whose primary purpose is to ensure that teaching activities run smoothly, also serve to support teaching and improve educational outcomes. It is the first and most important step in constructing purposeful resources. The primary goal of educational resource development is to promote educational reform, ensure that education meets the needs of modern society, and improve educational quality. The ultimate goal is to serve students, teachers, and other educators, so educational needs should be fully considered in both content and function. The collection, organisation, management, and dissemination of educational resources are all included in the educational resource service. The collection of educational resources is the first step in the digitalization of educational resources, as well as the processing and processing of resources. The organisation of educational resources, or the ordering of educational resources, is the process of transforming disorder into order by describing and ordering the external and content characteristics of resources using scientific rules and methods. Macromanagement and micromanagement are two types of educational resource management. The creation of music teaching information resources actually provides us with a three-dimensional, comprehensive, and high-efficiency music learning environment and resources, which includes music learning, music appreciation, music creation, and music communication. Students can learn according to their own preferences, including content, time, teaching methods, learning locations, and even tutors. The structure of the music education resource system constructed in this study is shown in Figure 1.
Network music education resources can be represented in various media forms supported by digital technology, such as audio, video, images, flash, and text. As a result, there is a wide range of music education resources available online. The definition of grades, types of resources, and file formats should be based on unified standards and conform to national norms, and the construction of educational resources must adhere to the laws and characteristics of education. The network music education resources cover a wide range of knowledge and content from various disciplines, demonstrating not only the depth and breadth of network resources but also their ability to spread advanced music teaching ideas and methods, as well as music education values and advanced music culture. Because music is a performing art, it necessitates the use of a large number of audio and video sources to convey its knowledge. There have been issues with the transmission of large amounts of music instructional resources over the Internet for a long time, such as slow speed and poor video quality. As a result, network and multimedia technology enable the transmission of music instructional resources and provide technical support for the creation of rich music instructional resources and the creation of a positive music learning environment.
It is not the accumulation of resource materials but the optimised resource integration according to the needs of teaching materials. Excellent resources not only provide teachers with discrete information and general information services but also provide teachers and students with higher-level information services, that is, knowledge services. Traditional educational resources information services include online inquiry and retrieval. With the development of network technology and the continuous improvement of users’ needs, people begin to try to carry out some special services such as information push service, targeted thematic service, and personalized customization service. The construction of educational resources can have four meanings: construction of material education resources: it is the foundation of the construction of educational resources and the key and core to be standardised; network course construction; evaluation of resource construction: it is the evaluation and screening of resources, and it needs to standardise the evaluation standards; and development of the educational resource management system. The specific contents of resources are ever changing, and their forms are different, so the corresponding management system must adapt to this change and make full use of their characteristics. Based on the network, network technology provides a powerful technical guarantee for discussion, communication, interaction, and cooperation between music students and instructors. Furthermore, the interaction, communication, and collaboration of online music education resources can effectively break the geographical space barrier, emphasising the interactive and convenient nature of online music education information resource exchange. Educational resource construction is a systematic project, in which the design, development, management, application, and evaluation of educational resources are all interconnected and interdependent steps. Because learners differ in their learning starting point, learning style, learning desire, learning pace, and so on, we should consider the resource library’s multilevel, diversified, open, and dynamic updating timeliness when building it, so that different learners can get the resources they need. The importance of resource distribution in resource construction is reflected in the purpose, dynamics, timeliness, and individuality of educational resource construction.
3.2. Construction of Music Instructional Resource Platform Based on Fuzzy Clustering Algorithm
Fuzzy clustering is a fuzzy mathematics method for classifying target objects using similarity relations. Clustering can be used on the Internet to classify documents and users’ access patterns in order to help users find information. Based on the fuzzy clustering algorithm, this study creates a music instructional resource platform. There are two server clusters in this study’s system deployment architecture: a web server cluster and a database server cluster. Through the global load balancer, the web server cluster performs session sharing and request distribution. The database is split into two parts: the master and slave servers. HDFS is one of Hadoop’s core technologies. The HDFS architecture is a master/slave. NameNode is in charge of managing the file system’s metadata and monitoring the state of DataNode, which stores the actual data. Following service verification, the system analyzes user access requests against the user model and generates a user agent to represent the user’s request. The system then divides the natural sentences entered by users intelligently based on the information carried by the user agent. Administrators and users can search for specific resources using keywords or categories. Users can see the queried resources by default based on the date they were uploaded. Users and resource administrators can sort resources manually as well. The business process of this system is shown in Figure 2.
When using Mahout to implement a clustering algorithm, it is required that the data model “vector” be used to uniformly describe the data, which is also known as data vectorization. This makes comparing the similarities between two data objects much easier. The features of related documents in the instructional resource database are extracted to form a number of related feature keywords, which are then analyzed using the analysis agent. Due to the constant growth of instructional resources, it may be necessary to add more storage devices to expand the existing storage system, so data storage must be scalable. The mobile terminal has flexible use characteristics but limited data storage and processing capabilities throughout the service mode. Therefore, it is only responsible for the input and output of data and transfers these complex functions to the server. All the data of users are stored in the cloud, while the cloud services and application processing of data that need to be obtained can be transmitted through the network. Because the fuzzy conditions of fuzzy rules are fuzzy propositions, the matching evidence is also fuzzy propositions. Therefore, the fuzzy matching method is needed to calculate the similarity of two fuzzy propositions, that is, the matching degree. Membership function is to use fuzzy sets to explain and analyze a certain fuzzy phenomenon, that is, to describe the degree, to which an object element belongs to a certain set through membership function. The proper construction of membership function is the key to express fuzzy concepts.
In order to better calculate the correlation between resources, this study introduces the concept of semantic distance, which is used to represent the force between concepts. The semantic distance between two semantic locations (such as formula (1)) in factor space is calculated using Euclidean distance, and the calculation formula is as shown in formula (2):
Based on the above definitions, the concepts in educational resource information are analyzed. If is the number of samples in the th cluster , then is the mean of the samples in the th cluster, namely,
The sum of squared errors between the sample in and the mean is added to all classes as follows:
In this study, TFIDF weight is used to measure the feature frequency of feature items appearing in documents. The formula for TFIDF is as follows:
Among them, refers to the word frequency, indicating the frequency of the word entry in the text . represents the total number of documents; represents the frequency of documents containing the feature word . Considering the length of the text to normalize the weights of the feature items, the formula is as follows:
A method for calculating the degree of correspondence between a single condition and a single piece of evidence, as well as the degree of correspondence between each subcondition in the combination of rules and the corresponding evidence, is selected. Different applications and users are allowed to be assigned to different NameNodes without affecting each other, thus improving performance; data blocks are stored in DataNodes. The system performs intelligent retrieval in the instructional resource database, obtaining a preliminary retrieval set of instructional resources, using a fuzzy proximity algorithm based on threshold and retrieval rules based on self-learning. This study systematically collects both teacher-provided instructional resources and instructional resources generated by students’ learning processes. By collecting and mining these data, teachers can better understand students’ learning behaviors, improve teaching methods, and further improve the quality of education and teaching. In this study, the uncertainty caused by randomness is expressed by credibility CF, while the uncertainty caused by fuzziness is still expressed by fuzziness. In addition, when the precondition of knowledge is the combination of multiple subconditions, the degree of influence of each subcondition on the conclusion is not exactly the same. Therefore, weighting factors are introduced into the antecedent conditions, and corresponding weights are given to each subcondition to illustrate their importance to the conclusion. Assuming that the educational resource contains semantic features, the corresponding weights of these semantic features are as follows:
The semantic position of educational resource in factor space is calculated by the following formula:
The Euclidean distance is used to calculate the distance between the online education resource and the online education resource . The calculation formula is as follows:
Among them, and are the semantic positions of two online educational resources and , is the dimension of the factor space, and is the value of resource on the dimension. The formula for calculating similarity is as follows:
Among them, is the interest preference value of user to the fuzzy cluster of instructional resources; is the average interest preference value of user to all fuzzy clusters of instructional resources. User ’s representation of instructional resources is consistent with user .
Learning agent gathers information about users’ online behavior at the client, such as copying, downloading, printing, and browsing time on resource pages, in order to mine users’ interests and preferences and update the user model accordingly. Users can browse all types of online resources after entering the platform, and the system will automatically generate some excellent resources, popular resources, new resources, and so on. The resource can be evaluated and downloaded after it has been opened. The server allows users to upload their own experimental reports, study reports, courseware, lesson plans, study guidance, and so on. Users can choose different vector implementation classes based on their needs when implementing the algorithm. The specific choice is determined by the algorithm’s characteristics. A personalized model of users is created by calculating users’ interest in instructional resources, and users with similar preferences are clustered using clustering technology. Users can tag resources by entering keyword tags or agreeing with the tags of other users. Users can tag resources while evaluating them or after they have completed the evaluation.
4. Result Analysis and Discussion
The music resource management system mainly collects the instructional resources generated in the process of music teaching and provides convenience for teachers’ teaching and students’ autonomous learning. In order to verify the effectiveness and system performance of the music instructional resource management system proposed in this study, this section makes an experimental analysis. The experimental data set adopts a university education and instructional resource database. The content covers nearly 5 TB of English, with more than 5,000 registered users and more than 800 daily visits. Through sorting, 26,719 pieces of interest evaluation data of 500 users on 1,000 instructional resources were collected. The higher the value, the higher the user’s interest in this instructional resource. About 70% of this data set is used as training data set and 30% as test data set. The evaluation results of users’ music instructional resources are listed in Table 1.
Mobile terminals play an important role in people’s daily life and work because of their high flexibility and practicality. Users can access remote servers by using their mobile terminal devices and mobile Internet connections and then get services such as underlying infrastructure, platform environment, cloud applications, and data storage and processing according to their own needs. The system includes more than 10 subfunctions, namely, resource management, resource download, resource evaluation, resource retrieval, user evaluation management, user information maintenance, and user authority management. In order to verify the performance of the algorithm, this section tests the recall rate and accuracy rate of the algorithm, respectively, and compares the results with those of different algorithms. Figure 3 shows the recall results of different algorithms. Figure 4 shows the accuracy results of different algorithms.
In this study, the platform uses the HBase database to ensure the fast and effective writing operation of unstructured instructional resource data; when data is queried, HQL statements are used for retrieval, which ensures the oneness of data flow and optimises the overall running efficiency of the platform system. The average absolute errors of the calculation methods are compared, and the experimental results are shown in Figure 5.
From the experimental results, it can be seen that the fuzzy clustering algorithm has a smaller MAE value than the traditional collaborative filtering algorithm and decision tree algorithm. This shows that the accuracy and quality of this algorithm are better. The resource classification function is mainly displayed through the tree structure, which coincides with the data structure of resource classification. Category administrators can search categories directly through the search box; we can also select the category to be operated by expanding the category tree. After selecting a category, we can modify, create new subcategories, delete, and so on. Experimenting again, the retrieval accuracy of different instructional resource management systems is shown in Figure 6.
When processing text objects, the system will convert data into vectors, use the high dimensions of vectors to represent the features of the text, and then classify them by the clustering algorithm to achieve accurate and efficient retrieval results. In this study, 10 experiments were conducted, respectively, and the data of indexes in this study are given in Table 2.
From the above experimental data, it can be clearly seen that the personalized resource retrieval service in this system has relatively high indexes. It has certain superior performance. In the music instructional resource platform designed in this study, Mahout, a clustering algorithm, is used to realize the retrieval of instructional resources by combining the vector space model. The vector model vectorizes the text data. It uses the high dimension of vector to represent the features in text information and converts the similarity between texts into the similarity between calculated vectors, which is usually measured by cosine distance. To further verify the feasibility of the system, this study compares the stability of different instructional resource management systems, and the results are shown in Figure 7.
From the above data analysis, it can be seen that the stability of this music instructional resource management system is good, and its stability is higher than other comparison systems. The stability of this system can reach 95.37%; the retrieval accuracy is as high as 96.24%, which is 8.94% higher than the traditional method. The functions of the music instructional resource management system constructed in this study fully meet the requirements, and its performance in all aspects is good; compared with the traditional system, this system effectively improves the satisfaction of resource users.
Informatization views information as a fundamental component of the educational system and employs modern information technology in the field of education to support educational reform and development. Educational resources are constantly increasing as educational informatization progresses. The creation of educational resources has become a critical aspect of educational informatization. A music resource management system is proposed in the context of educational informatization. The music resource management system organises educational resources into clusters, resulting in a more efficient and comprehensive resource organisation. Existing educational resources should be integrated and managed to some extent. Moreover, the music instructional resource management system proposed in this study has a flexible architecture, high reliability, and high reading and writing performance; the core instructional resource management function has been fully realized in terms of business requirements. The system uses a personalized service mechanism. On the one hand, it effectively realizes new users’ personalized initial resource selection; on the other hand, it effectively realizes old users’ personalized update resource selection. Many experiments show that this system’s stability can reach 95.37%. The retrieval accuracy is up to 96.24%, which is 8.94% better than the traditional method. The music instructional resource management system developed in this study fully meets the requirements. It can effectively provide high-quality resources for music teaching and better meet the user experience effect. Although this study provides a more comprehensive and in-depth understanding of the storage and processing of instructional resources on the platform using various technologies, some issues remain. The system will be improved and perfected in the future to better serve educators and students.
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The author declares that there are no conflicts of interest.
C. Long and S. Wang, “Music classroom assistant teaching system based on intelligent speech recognition,” Journal of Intelligent and Fuzzy Systems, vol. 1, pp. 1–10.View at: Google Scholar
W. Jennifer, “A pile of drums: putting theory into practice in culturally diverse music education,” International Journal of Music Education, vol. 38, no. 1, pp. 79–92, 2019.View at: Google Scholar
O. Zandén and C. F. Thorgersen, “Teaching for learning or teaching for documentation? Music teachers’ perspectives on a Swedish curriculum reform,” British Journal of Music Education, vol. 32, no. 1, pp. 37–50, 2015.View at: Google Scholar
Á. Arnaiz-González, J.-F. Díez-Pastor, I. Ramos-Pérez, and C. García-Osorio, “Seshat - a web-based educational resource for teaching the most common algorithms of lexical analysis,” Computer Applications in Engineering Education, vol. 26, no. 6, pp. 2255–2265, 2018.View at: Publisher Site | Google Scholar