A Time-based Recommendation System using a Sequential Pattern Matching Method
Recommender systems have significant applications in both industry and academia. Collaborative filtering
methods are the most widely used in industrial applications. These algorithms utilize preferences of similar users to provide
suggestions for a target user. Although the users’ preferences may vary over time, traditional collaborative filtering
algorithms fail to consider this important issue. Sequential pattern of ratings is another important factor which disregarded by
collaborative filtering. In this paper, a novel recommendation method is proposed based on a sequential pattern matching
approach. The proposed method takes into account the time of ratings to calculate similarity values. Experimental results on
a benchmark dataset show that the proposed method significantly outperforms other recommendation methods.
Index Terms - Recommendation System, Sequential Pattern Matching, Similarity Measure, Time.