Hybrid Recommendation Using Association Rule Mining by Partial Evaluation of Web Personalization for Retrieval Effectiveness

World Wide Web is the biggest source of information. Though the World Wide Web contains a tremendous amount of data, most of the data is irrelevant and inaccurate from users’ point of view. Consequently it has become increasingly necessary for users to utilize automated tools such as recommender systems in order to discover, extract, filter, and evaluate the desired information and resources. Most recommendation algorithms attempt to alleviate information overload by identifying which items a user will find worthwhile. Web page recommender systems predict the information needs of users and provide them with recommendations to facilitate their navigation. Web content and Web usage mining techniques are employed as conventional methods for recommendation. In this paper, we proposed hybrid recommendation systems in m-commerce, which could integrate multiple association rules together to improve recommendation performance. The effects of the hybrid recommenders are examined by comparing the results of hybrid system against the results of single recommendation method. Result shows that the hybrid recommender provides successful recommendation when the recommended page is generated by all the systems of the hybrid. Our proposed approach based on both straight and meandering rules are attached into one set of global association rules, which might be used for the recommendation of web pages and for personalization. These experiments have shown that the use of reduced datasets saves computational time, and neighbor information improves performance.