In view of the limitations of traditional symbolic aggregate approximation which easily neglect the local morphological character of time series when applied to feature representation, and the advantages of dynamic time warping in the similarity measurement, a similarity measure based on the numeric symbolic and shape feature is proposed. Through representing based on the numeric symbolic and shape feature, a distance measure combined with dynamic time warping and symbolic distance measure is proposed, which has a better reflection of data distribution and morphologic characteristic. Experimental results show that the proposed method has better effect of classification and certain superiority.