Mining patterns in long sequential data with noise
Wei Wang, Jiong Yang, Philip S. Yu · ACM SIGKDD Explorations Newsletter · 2000
Pattern discovery in time series data has been a problem of great importance in many fields, e.g., computational biology, performance analysis, consumer behavior, etc. Recently, considerable amount of research has been carried out in this area. The facts that the input data is typically very large and noises may present in various formats pose great challenge to the mining process. Recently, we have made several new research advances in this area. In this paper, we present some of them. We will survey new models proposed to address different types of noises as well as scalable algorithms developed for efficiently mining patterns under each model. 1.