An efficient and accurate recommendation strategy using degree classification criteria for item-based collaborative filtering

Junpeng Guo, Jiangzhou Deng, Xun Ran, Yong Wang, Hang Jin

Abstract: Highlights•An extended classification criteria are proposed to assign items to more classes.•Hellinger distance based item similarity is proposed to evaluate similarities.•A sigmoid function is used to emphasize the importance of the co-rated items.•Results reveal that our algorithm has a favorable efficiency and accuracy.