A sequential pattern mining algorithm using rough set theory

https://doi.org/10.1016/j.ijar.2011.03.002Get rights and content
Under an Elsevier user license
open archive

Abstract

Sequential pattern mining is a crucial but challenging task in many applications, e.g., analyzing the behaviors of data in transactions and discovering frequent patterns in time series data. This task becomes difficult when valuable patterns are locally or implicitly involved in noisy data. In this paper, we propose a method for mining such local patterns from sequences. Using rough set theory, we describe an algorithm for generating decision rules that take into account local patterns for arriving at a particular decision. To apply sequential data to rough set theory, the size of local patterns is specified, allowing a set of sequences to be transformed into a sequential information system. We use the discernibility of decision classes to establish evaluation criteria for the decision rules in the sequential information system.

Keywords

Rough set theory
Sequential pattern mining
Local patterns

Cited by (0)

This paper is an extended version of [12].