SEQUENTIAL DATA MINING: EXPLORING THE REDUNDANT PATTERNS FROM SEQUENCES TO MINIMIZE THE OVERALL PATTERNS

Recent studies in discovering patterns from sequence data have shown the significant impact in many aspects of data mining. In this research, a novel method of finding the redundant pattern is proposed. To efficiently discover the redundant pattern, the focus is on developing new algorithms. Rapid increase of the sequential data has created the problem of discovering meaningful patterns from sequences. The most challenging problem is to find repeating patterns with gap constraints. In this work, we identify a new research for mining the redundant patterns with gap constraints. To solve the problem, we propose algorithm with components such as: (1) Data-driven pattern generation approach to avoid generating unnecessary candidates for validation. (2) Back-tracking pattern search process to discover approximate occurrences of a pattern under user specified gap constraints. (3) An Apriori-like deterministic pruning approach to progressively prune patterns and cease the search process if necessary. It is proposed to conduct experimental analysis on the synthetic and standard data sets. It is also proposed to conduct comparative analysis of the developed algorithms with the state of art algorithms.