Applying Metaheuristic-based Boosted Incremental K-Nearest-Neighborhood Based to Collaborative Filtering Method for Recommendation Systems

• Main Authors: You-Liang - Li, 李宥良
• Other Authors: Ren-Jieh Kuo
• Format: Others
• Language: en_US
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/39477445371045141211

碩士 === 國立臺灣科技大學 === 工業管理系 === 105 === Global e-commerce has grown very fast, and daily revenue can be up to billion US dollars. Many industries follow the trend and earn lots of money, such as: Amazon and Taobao. To raise revenue, most of e-commerce companies endeavor to develop recommendation system in order to find out potential customers or stick customers. Recommendation systems can be implemented by some methods and the most well-known method is collaborative filtering. It mainly uses similar user’s records to recommend what similar users like. Its advantage is no need to analyze the product’s profile. This study uses the boosting the K-nearest-neighbors based incremental collaborative filtering method (BIKNN) as collaborative filtering, and uses metaheuristics to optimize BIKNN’s parameters to improve prediction performance. Besides, this study uses batch updating method instead of incremental updating method to reduce the computational time. To validate the proposed algorithm, this study conducts K-fold cross validation. Four benchmark dataset are used in the experiment: Movielens 100K, Movielens 1M, Books-crossing and Restaurants. The experimental results indicate that the three metaheuristic-based BIKNN algorithms are different and better than basic KNN and original BIKNN.