Collaborative Similarity Embedding for Recommender Systems

17 Feb 2019  ·  Chih-Ming Chen, Chuan-Ju Wang, Ming-Feng Tsai, Yi-Hsuan Yang ·

We present collaborative similarity embedding (CSE), a unified framework that exploits comprehensive collaborative relations available in a user-item bipartite graph for representation learning and recommendation. In the proposed framework, we differentiate two types of proximity relations: direct proximity and k-th order neighborhood proximity. While learning from the former exploits direct user-item associations observable from the graph, learning from the latter makes use of implicit associations such as user-user similarities and item-item similarities, which can provide valuable information especially when the graph is sparse. Moreover, for improving scalability and flexibility, we propose a sampling technique that is specifically designed to capture the two types of proximity relations. Extensive experiments on eight benchmark datasets show that CSE yields significantly better performance than state-of-the-art recommendation methods.

PDF Abstract

Results from the Paper

 Ranked #1 on Recommendation Systems on Netflix (mAP@10 metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Recommendation Systems CiteULike RATE-CSE Recall@10 0.2362 # 1
mAP@10 0.1452 # 1
Recommendation Systems Echonest RANK-CSE Recall@10 0.1358 # 1
mAP@10 0.0679 # 1
Recommendation Systems Epinions-Extend RANK-CSE Recall@10 0.1767 # 1
mAP@10 0.0921 # 1
Recommendation Systems Frappe RATE-CSE Recall@10 33.47 # 1
mAP@10 0.2047 # 1
Recommendation Systems Last.FM-360k RANK-CSE Recall@10 0.1762 # 1
mAP@10 0.097 # 1
Recommendation Systems MovieLens-Latest RATE-CSE Recall@10 0.3225 # 1
mAP@10 0.199 # 1
Recommendation Systems Netflix RATE-CSE Recall@10 0.2014 # 4
mAP@10 0.1039 # 1


No methods listed for this paper. Add relevant methods here