Hyo-Sang Lim  Juwon Lee

10.5626/JOK.2018.45.4.403

SET Set Similarity set-based similar sequence matching Optimization

The set-based similar sequence matching method involves searching for data set sequences that are similar to a query set sequence. In the method, the similarity between two sets is represented as the size of intersection between them. However, there is a critical performances issue for calculating intersection size if the number of sets is large. In the past, authors of the present work proposed a method to improve the performance of set-based similar sequence matching using simple index structure. In this paper, we propose an optimization method for more efficient running of set-based similar sequence matching. Our method is based on pruning that excludes unnecessary calculation. Through experiments, we show that the proposed method reduces the execution time by about 20% compared to the existing methods.