Design and application research of traditional Chinese medicine teaching resource recommendation system based on multivariate data mining driven algorithm

Abstract

With the wide spread of traditional Chinese medicine culture around the world and the rapid development of modern information technology, the demand for traditional Chinese medicine teaching resources is increasing, and how to achieve personalized and accurate teaching resource recommendation has become the focus of much research. At present, there are many problems that need to be solved urgently: first, resources are scattered and distributed on various platforms, due to the lack of a unified integration platform, teachers and students need to frequently switch to multiple channels when searching for resources, and the average time spent on each resource search is about 30 min; Second, the quality of resources is uneven, there are different levels of content accuracy, and some resources even contain incorrect or outdated information, which seriously affects the learning effect. Third, the traditional recommendation method is difficult to meet the personalized needs of users, resulting in low resource utilization. Therefore, it is necessary to design and apply an efficient and intelligent recommendation system to optimize the allocation of TCM teaching resources and improve the teaching effect. This study focuses on multivariate data mining-driven algorithms, and is committed to the design and application of a TCM teaching resource recommendation system. The study uses a hybrid recommendation algorithm, combines collaborative filtering with content filtering, and integrates it into the knowledge graph of traditional Chinese medicine to realize in-depth data mining. Experimental data show that the recommended system has excellent performance in accuracy, recall and F1 value, reaching 85.6 %, 79.2 % and 82.3 %, respectively, which is significantly improved compared with the traditional recommendation system built by standard collaborative filtering algorithm (its F1 value is only 68.5 %). In the user satisfaction survey, >90 % of users agreed that the recommendation results were effective in meeting personalized learning needs. Practical application shows that the system can reduce the average time of teachers' resource search by 40 %, and improve the learning efficiency of students by about 35 % due to the acquisition of accurate resources, which not only effectively improves the utilization efficiency of TCM teaching resources, but also provides important support for the modernization and intelligent development of TCM education. In this study, a multivariate data mining-driven algorithm was used to design a teaching resource recommendation system. Learners can obtain personalized learning resources and improve their practical skills. Teachers are able to optimize teaching content and evaluation; Managers can achieve scientific decision-making and quality monitoring, ultimately promote educational equity, activate characteristic resources, and promote the digital transformation of traditional Chinese medicine education and the improvement of the quality of talent training.

Keywords

• Chinese medicine teaching resources
• Recommendation systems
• Multivariate data mining
• Hybrid recommendation algorithm
• Personalized learning