A Multidimensional Mining and Pattern Recognition Approach for Piano Teaching Behavior Data in Music Education

Digital technology assistance has become a new trend in the development of education, the feedback of classroom teacher behavior as a service of teaching process evaluation is an important channel for teachers to improve their teaching ability. This paper is oriented towards piano teaching and mines teacher behavior data using multidimensional data mining. Then the Teacher-Set IE algorithm is proposed for the problem of interfering factors for teaching behavior identification and extraction, and finally the teacher behavior pattern recognition model (3D BP-TBR) is constructed based on three-dimensional bilinear pooling. This paper investigates the learning efficiency and learning gain pattern under different teaching behavior patterns, and examines the performance of the teacher behavior recognition model. The learning efficiency and learning gains under the rhythmic teaching behavior pattern in the whole course learning are relatively better compared to other teaching behavior patterns. On the other hand, in the identification of teaching behavior patterns, the average identification precision, recall and F1 value of the four teaching behavior patterns, namely, stagnant, focused, rushed and rhythmic, are 97.31%, 96.96% and 97.34%, respectively, which indicates that the pattern recognition method in this paper is effective and can adequately identify the piano teaching behavior patterns. This paper’s research on teachers’ teaching behavior pattern data mining rows and recognition is for the valuable information behind the educational data to promote teachers for the optimization of the teaching process, teaching results and teaching environment.