Individualized Recommender System Using Keywords With Efficient Similarity Computation Employing Map Reduce

Abstract

The advancement in E-commerce enabled the websites to overwhelm  users with numerous services. Often times, users find it a challenging task to choose the appropriate and best service from the availabe services. To provide the user with the most appropriate options, Recommender system, which is an information filtering technique, can be employed.  Traditional recommender systems present the same ratings and rankings of services to different users without considering diverse users' preferences. A individualized recommender system addresses the versatile  requirements of different users. Specifically, keywords are used to indicate users' preferences, and a user-based Collaborative Filtering algorithm is adopted to generate appropriate recommendations. However, finding similarity, which is heart of any recommender system, plays a vital role. The best similarity computation method contributes massively for an efficient recommender system.

The proposed work implements three similarity computation methods. These methods are compared with the existing methods and the results proved that the proposed techniques outperformed the existing techniques. To improve its scalability and efficiency in big data environment, proposed approach is implemented on Hadoop, a widely-adopted distributed computing platform using the MapReduce parallel processing paradigm.