Mining colossal patterns with length constraints

Mining of colossal patterns is used to mine patterns in databases with many attributes and values, but the number of instances in each database is small. Although many efficient approaches for extracting colossal patterns have been proposed, they cannot be applied to colossal pattern mining with constraints. In this paper, we solve the challenge of extracting colossal patterns with length constraints. Firstly, we describe the problems of min-length constraint and max-length constraint and combine them with length constraints. After that, we evolve a proposal for efficiently truncating candidates in the mining process and another one for fast checking of candidates. Based on these properties, we offer the mining algorithm of Length Constraints for Colossal Pattern (LCCP) to extract colossal patterns with length constraints. Experiments are also conducted to show the effectiveness of the proposed LCCP algorithm with a comparison to some other ones.Web of Scienc