Data mining is the process of discovering new relevant information in terms of patterns from large amount of data. Association rule mining is one of very important data mining techniques. Swarm optimization is a new subfield of artificial intelligence which studies the cooperative performance of simple agents. In this paper, proposed a new efficient algorithm for exploring high-class association rules by Particle Swarm Optimization (PSO) algorithm. The proposed method mine interesting and understandable association rules without using the minimum support and the minimum confidence thresholds in only single scan.