The Performance Evaluation of Recommendation Algorithm Using Mahout Framework
ZiDong Yan, Qingchun Hu, Chengyu Tang, Kelang Yang, Hongyu Cai · 2020
In order to achieve a better recommendation effect, the optimization and improvement of the recommendation algorithm has been the research hotspot of the recommendation system. Similarity is the core problem of recommendation algorithm, so in this paper, a novel method of calculating similarity in collaborative filtering recommendation was proposed to make the recommendation better. We used the weighted average method to combine various similarity algorithms on the calculation of similarity, so as to improve the accuracy of recommend results and the stability of the algorithm. In order to test the weighted coefficient of similarity and tuning, the experiment is conducted on open source data sets and Mahout framework. Finally, an effective way to improve the collaborative filtering algorithm is presented.