A novel KNN approach for session-based recommendation

Abstract

The KNN approach, which is widely used in recommender systems because of its efficiency, robustness and interpretability, is proposed for session-based recommendation recently and outperforms recurrent neural network algorithms. It captures the most recent co-occurrence information of items by considering the interaction time. However, it neglects the co-occurrence information of items in the historical behavior which is interacted earlier than others and cannot discriminate the impact of vertices with different popularity. Due to these observations, this paper presents a novel KNN approach to address these issues for session-based recommendation. Specifically, a diffusion-based similarity method is proposed for incorporating the popularity of items, and the candidate selection method is proposed to capture more co-occurrence information of items in the same session efficiently. Comprehensive experiments are conducted to demonstrate the effectiveness of our KNN approach over the state-of-the-art KNN approach for session-based recommendation on three benchmark datasets.