Abstract: Because the teaching effect of online teaching is influenced by teaching means and teaching content, the accuracy of online teaching effect evaluation of physical education is low. Therefore, a multi-index evaluation method based on fuzzy K-means clustering algorithm is designed. Firstly, the multi-index evaluation model of online sports teaching effect is constructed, the key elements that affect online sports teaching quality are divided, the index weights are solved, and the evaluation index parameters of online sports teaching effect are clustered and merged by fuzzy K-means clustering method, so that the quantitative recursive results can be obtained after the parameters of convergence index with nonlinear time series characteristics are constrained, avoiding the interference of teaching environment and completing the high-precision online sports teaching effect evaluation. The experimental results show that the proposed evaluation method can improve the operation speed on the basis of ensuring the high-precision evaluation of online teaching effect of physical education.

Key words: fuzzy K-means clustering; , multi index evaluation; , physical education; , online teaching; , teaching quality