CCR

Previous Issue

Asian Journal of Information and Communications 10(2)
Asian Journal of Information and Communications 10(2)

Call for Paper

Add to Bookmark

Editorial Office

Asian Journal of Information and Communications

Similarity search for localized patterns in time series data Ayush Rathi and Durga Toshniwal

  • Page:58-69
  • 2019-05-02

Time series data are of crucial importance as they depict trend of various entities over time. In this paper we give an approach to find similarity in time series based on angular differences and increasing decreasing patterns. The angular differences between two time series are computed by taking differences of their respective angles, where angles are the measured between the consecutive lines connecting the time series values. Differences between corresponding increasing-decreasing patterns are also taken into consideration and a method is proposed to find similarity based on these features. To demonstrate the effectiveness of the proposed approach, it is applied to various different datasets including synthetic, US retail store and Indian rainfall datasets. The comparative analysis of the experimental results of the proposed approach with real life scenarios show that the approach can be efficiently utilized to find similarity in time series for a wide range of applications.

Keywords: time-series, similarity, angular measure, increase-decrease pattern

  PDF Download
  • Facebook
  • Share onTwitter