Multivariate time series classification is a key problem in many fields, but the current research on multivariate time series classification is faced with some problems, such as high dimensionality of original data, low accuracy, and lack of interpretability, which limits the performance improvement of models and makes it difficult to meet the actual requirements. Aim at above problem, a multivariate time series classification method based on Shapelets is proposed. Firstly, unsupervised Shapelet learning of adaptive neighbors is used to automatically learn significant multivariate Shapelets by combining Shapelets transform and adaptive weights. Then, the method is combined with Shapelet similarity and class label constraint to enhance the interpretability and classification accuracy of the model. Finally, the optimization strategy of the model is proposed to obtain the best Shapelets to further improve the classification accuracy of the model. In this paper, three different types of 11 algorithms are compared on 11 public data sets, and the experimental results show that the proposed algorithm has high classification accuracy.