Hybrid Recommendation Algorithm Based on MapReduce and Its Application

Cheng LI,Han CAO,Jun SHI
DOI: https://doi.org/10.3969/j.issn.1673-629X.2016.04.016
2016-01-01
Abstract:For the shortcomings of traditional project-based and user-based collaborative filtering algorithm,a new parallel recommenda-tion algorithm is proposed,combined user-based with project-based collaborative filtering algorithm and improved them. Through Ma-pReduce compilation,the new algorithm can run in Hadoop cloud platform. To guarantee the personalized recommendation,it can take ad-vantages of the collaborative filtering algorithm based on user defined a certain number of neighbors. At the same time,the project-based collaborative filtering algorithm is used to recommend. Finally,according to the comprehensive adjusted score prediction method,the rec-ommended results are obtained. The experimental results show that the algorithm becomes more superior in the case of a large number of processing speed,and improves the accuracy of recommendation. Simultaneously,the algorithm is applied in local tourism of Xi’ an refer-ral service for several major attractions to recommend. The accuracy of the new algorithm has been verified in practical applications.
What problem does this paper attempt to address?