Sequential Pattern Mining for e-Commerce Recommender System
Ridho Trivonanda, Rahmad Mahendra, Indra Darmawan Budi, Rani Aulia Hidayat · 2020
Recommender system is one of the strategies carried out by e-commerce to increase their users' satisfaction. In this paper, we implement sequence pattern mining for recommender system in e-commerce domain. We perform the PrefixSpan algorithm to mine the frequent patterns. The frequent patterns generate the suitable rules for the recommender system. The experiments reported in this paper utilize the datasets from two different marketplaces in Indonesia. The best performance of recommendation model is obtained when applying the rules derived from the product category with minimum confidence of 0.5.