Discovering Approximate Time Series Motif Based on MP_C Method with the Support of Skyline Index

Abstract

Time series motifs are frequently occurring but unknown sequences in a time series database or subsequences of a longer time series. Discovering time series motifs is a crucial task in time series data mining. Among a dozen algorithms have been proposed for discovering time series motifs, the most popular algorithm is Random Projection. This algorithm can find motifs in linear time. However, it still has some drawbacks. In this paper, we propose a novel method for discovering approximate motifs in time series. This method is based on MP_C dimensionality reduction method with the support of Skyline Index. Our method is disk-efficient because it only needs a single scan over the entire time series database. The experimental results showed that our proposed algorithm outperforms Random Projection in efficiency.