[2S6-IS-3d] Machine learning

[2S6-IS-3d-01] A study of sequence matching method considering data transition

〇Masashi Kamura¹, Hiroyuki Kasai¹ (1. Waseda University )

Regular

Keywords:machine learning, sequence matching, optimal transport

In this paper, we focus on the problem of measuring the distance between sequences whose order has some meaning. The existing method, Order-Preserving Wasserstein distance (OPW), has a problem that it does not take into account the similarity and neighbor relationship of the elements too much. In this paper, we propose a method that considers the neighbor relationship between elements by using the transitions of elements in addition to OPW matching. From numerical evaluation experiments, we show that the proposed method improves the classification accuracy for some data sets.