Personal Recommender System Based on Agglomerative Clustering together with User-based and Item-based Collaborative Filtering Methods

摘要

The objective of this study is to develop a Personal Integrated Recommender System. The Recommender System plays an important role and is crucial to our everyday lives in online shopping, meanwhile, it also encounters various problems e.g. scalable data, data sparsity, data accuracy, and having a lot of new users. Therefore, new techniques have been introduced and integrated with the recommender system in order to solve the problems and improve for greater recommender system efficiency. This study, an Agglomerative Clustering together with a User-base and Item-base Collaborative Filtering Method is proposed. By combining the strengths of each method, we can improve the recommender system efficiency and accuracy. The results show that the system being developed generates better values of the area under the curve, precision, normalized discounted cumulative gain, and mean average precision than using only User-based Collaborative Filtering or Item-based Collaborative Filtering alone. Therefore, we can conclude that the Personal Recommender System developed based on Agglomerative Clustering together with User-based and Item-based Collaborative Filtering Method has the ability to increase system efficiency and is applicable. When modern technology arrives in the future, it may reduce the processing time and increase precision.