A PIP-based evolutionary approach for time series segmentation and pattern discovery
Hsieh-Hui Yu, Vincent S. M. Tseng, Chun-Hao Chen, Tzung‐Pei Hong · 2010
In the past, we proposed a time series segmentation approach by combining the clustering technique, the Discrete Wavelet Transformation (DWT) and the genetic algorithm to automatically find segments and patterns from a time series. In this paper, we propose a PIP-based evolutionary approach, which uses Perceptually Important Points (PIP) instead of DWT, to effectively adjust the length of subsequences for finding appropriate segments and patterns and avoiding some problems in the previous approach. For achieving the purpose, the enhanced suitability factor in the fitness function which is modified from the previous approach, is designed. Experimental results on a real financial dataset also show the effectiveness of the proposed approach.