A Study on Controlled-Sized Clustering for Time Series Data

Abstract

The analysis of time series data has been actively studied in various fields such as biology and economics. Clustering is a method that summarizes a set of objects into several subsets of objects based on similarity measures. It is necessary to deflne a suitable similarity between objects. In addition, it is also necessary to consider shift-invariance when dealing with time series data. k-Shape clustering is one of the representative clustering methods for time series data. It is known that k-Shape clustering is an algorithm that employs dissimilarities that satisfy several invariances. The dissimilarities used in k-Shape clustering is robust to differences in features of time series data. In this paper, the controlled-sized k-Shape clustering is proposed to to handle imbalanced data. Numerical experiments suggest that the proposed method does not show outstanding performance compared to k-Shape.