Tag-Aware Personalized Recommendation Using a Deep-Semantic Similarity Model with Negative Sampling

Conference item

With the rapid growth of social tagging systems, many efforts have been put on tag-aware personalized recommendation. However, due to uncontrolled vocabularies, social tags are usually redundant, sparse, and ambiguous. In this paper, we propose a deep neural network approach to solve this problem by mapping both the tag-based user and item profiles to an abstract deep feature space, where the deepsemantic similarities between users and their target items (resp., irrelevant items) are maximized (resp., minimized). Due to huge numbers of online items, the training of this model is usually computationally expensive in the real-world context. Therefore, we introduce negative sampling, which significantly increases the model’s training efficiency (109.6 times quicker) and ensures the scalability in practice. Experimental results show that our model can significantly outperform the state-of-the-art baselines in tag-aware personalized recommendation: e.g., its mean reciprocal rank is between 5.7 and 16.5 times better than the baselines.

Authors: Xu, Z; Chen, C; Lukasiewicz, T; Miao, Y; Meng, X

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Association for Computing Machinery
• Host title: CIKM 2016: 25th ACM International Conference on Information and Knowledge Management
• Journal: 25th ACM International Conference on Information and Knowledge Management More from this journal
• Publication date: 2016-01-01
• Acceptance date: 2016-07-19
• DOI: 10.1145/2983323.2983874
• Keywords: Negative Sampling; Deep Neural Network; Tag-Aware Personalized Recommendation; Deep-Semantic Similarity