Density-Based Clustering by P System with Active Membranes on Commodity Recommendation in E-commerce Websites

With the Popularity of shopping online in people's daily economic life, the commodity recommendation mechanism in e-commerce platform is presented to help customers quickly and accurately find the suitable product. Using the possibility of changing membrane structure, a variant of P system with active membranes is proposed to solve commodity recommendation problems. In this paper, the commodity recommendation problem is transformed into a density-based clustering problem firstly. Then it specifies the procedure of realizing this problem and a P system with a sequence of new rules is designed. The computation complexity of DBSCAN clustering algorithm in this system is , while the original DBSCAN clustering algorithm is without spatial query. This new model of P system can reduce the computation complexity of clustering process and improve the efficiency to solve the problems of commodity recommendation. Through example verification, this new model of P system is proved to be feasible and effective to achieve this practical issue. Key-Words: Membrane Computing; P System; Active Membranes; DBSCAN; Commodity Recommendation