Hybrid recommender system using systolic tree for pattern mining

Abstract

Recommender systems offer effective recommendations to the users whilst interacting with huge information spaces. By means of quality recommendations, the Recommender Systems have been able to enhance the user experience and thus, effectively handle the information overload issue. Sequential pattern mining is a technique of data mining used for the identification of the co-occurrence relationships by taking into account the order of transactions. This work will present the implementation of sequence pattern mining for recommender systems within the domain of e-commerce. This work will execute the Systolic tree algorithm for mining the frequent patterns to yield rules that are feasible for the recommender system. The Hybrid Recommender system will combine two or more of the aforementioned techniques to give its recommendations. Out of all the aforementioned system types, the Collaborative Filtering (CF) methods offer the most promising outcomes. With the utilization of the sequential pattern mining algorithm, efficient extraction of Frequent Patterns (FP) from the database is achieved. A candidate sub-sequence generation-and-test method is adopted in conventional sequential mining algorithms. However, since this approach will yield a huge candidate set, it is not ideal when a large amount of data is involved. Since the data is composed of numerous features, all of which may not have any relation with one another, the utilization of feature selection helps remove unrelated features from the data with minimal information loss. newline