Food Recommendation System Using Sequential Pattern Mining

Abstract

Sequential pattern mining is an important subfield in data mining. In this paper, we introduce a sequential pattern mining-based food recommendation system. The purchase history of users is analyzed to find their sequential patterns using SPADE algorithm [5]. These patterns are then used to predict the next possible purchase. One of these patterns will be shown as a special offer with discount. The proposed approach is experimented on real transaction data.

It demonstrate that the proposed system effectively improves the efficiency for mining sequential patterns, increases the user-relevance of the identified sequential patterns, and most importantly, generates significantly more accurate next-items recommendation for the target users.