ABSTRACT
Next item recommendation is a crucial task of session-based recommendation. However, the gap between the optimization objective (Binary Cross Entropy) and the ranking metric (Mean Reciprocal Rank) has not been well-explored, resulting in sub-optimal recommendations. In this paper, we propose a novel objective function, namely Adjusted-RR, to directly optimize Mean Reciprocal Rank. Specifically, Adjusted-RR adopts Bayesian Average to adjust Reciprocal Rank loss with Normal Rank loss by creating position-aware weights between them. Adjusted-RR is a plug-and-play objective that is compatible with various models. We apply Adjusted-RR on two base models and two datasets, and experimental results show that it makes a significant improvement in the next item recommendation.
Supplemental Material
- Suhrid Balakrishnan and Sumit Chopra. 2012. Collaborative ranking. In WSDM. ACM, 143--152. https://doi.org/10.1145/2124295.2124314Google ScholarDigital Library
- Andrew Brown, Weidi Xie, Vicky Kalogeiton, and Andrew Zisserman. 2020. Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval. In ECCV, Vol. 12354. 677--694. https://doi.org/10.1007/978-3-030-58545-7_39Google ScholarDigital Library
- Fatih Cakir, Kun He, Xide Xia, Brian Kulis, and Stan Sclaroff. 2019. Deep Metric Learning to Rank. In CVPR. 1861--1870. https://doi.org/10.1109/CVPR.2019.00196Google ScholarCross Ref
- Paul Henderson and Vittorio Ferrari. 2016. End-to-End Training of Object Class Detectors for Mean Average Precision. In ACCV, Vol. 10115. Springer, 198--213. https://doi.org/10.1007/978-3-319-54193-8_13Google ScholarCross Ref
- Alan Herschtal and Bhavani Raskutti. 2004. Optimising area under the ROC curve using gradient descent. In ICML, Vol. 69. ACM. https://doi.org/10.1145/1015330.1015366Google ScholarDigital Library
- Hang Li. 2011. A Short Introduction to Learning to Rank. IEICE Transactions, Vol. 94-D (10 2011), 1854--1862.Google ScholarCross Ref
- Pritish Mohapatra, Michal Rolínek, C. V. Jawahar, Vladimir Kolmogorov, and M. Pawan Kumar. 2018. Efficient Optimization for Rank-Based Loss Functions. In CVPR. Computer Vision Foundation / IEEE Computer Society, 3693--3701.Google Scholar
- Yitong Pang, Lingfei Wu, Qi Shen, Yiming Zhang, Zhihua Wei, Fangli Xu, Ethan Chang, Bo Long, and Jian Pei. 2022. Heterogeneous Global Graph Neural Networks for Personalized Session-based Recommendation. In WSDM. ACM, 775--783. https://doi.org/10.1145/3488560.3498505Google ScholarDigital Library
- Michal Rolinek, Vit Musil, Anselm Paulus, Marin Vlastelica, Claudio Michaelis, and Georg Martius. 2020. Optimizing Rank-Based Metrics With Blackbox Differentiation. In CVPR.Google Scholar
- Song-Qing Shen, Bin-Bin Yang, and Wei Gao. 2020. AUC Optimization with a Reject Option. In AAAI. AAAI Press, 5684--5691. https://ojs.aaai.org/index.php/AAAI/article/view/6023Google Scholar
- Yue Shi, Alexandros Karatzoglou, Linas Baltrunas, Martha A. Larson, Nuria Oliver, and Alan Hanjalic. 2012. CLiMF: learning to maximize reciprocal rank with collaborative less-is-more filtering. In RecSys. ACM, 139--146. https://doi.org/10.1145/2365952.2365981Google ScholarDigital Library
- Yue Shi, Martha A. Larson, and Alan Hanjalic. 2010. List-wise learning to rank with matrix factorization for collaborative filtering. In RecSys. ACM, 269--272. https://doi.org/10.1145/1864708.1864764Google ScholarDigital Library
- Hyun Oh Song, Yu Xiang, Stefanie Jegelka, and Silvio Savarese. 2016. Deep Metric Learning via Lifted Structured Feature Embedding. In CVPR. IEEE Computer Society, 4004--4012. https://doi.org/10.1109/CVPR.2016.434Google ScholarCross Ref
- Shuang Tang, Fangyuan Luo, and Jun Wu. 2022. Smooth-AUC: Smoothing the Path Towards Rank-based CTR Prediction. In SIGIR. ACM, 2400--2404. https://doi.org/10.1145/3477495.3531865Google ScholarDigital Library
- Shoujin Wang, Longbing Cao, Yan Wang, Quan Z. Sheng, Mehmet A. Orgun, and Defu Lian. 2022. A Survey on Session-based Recommender Systems. ACM Comput. Surv., Vol. 54, 7 (2022), 154:1--154:38. https://doi.org/10.1145/3465401Google ScholarDigital Library
- Ziyang Wang, Wei Wei, Gao Cong, Xiao-Li Li, Xianling Mao, and Minghui Qiu. 2020. Global Context Enhanced Graph Neural Networks for Session-based Recommendation. In SIGIR. ACM, 169--178. https://doi.org/10.1145/3397271.3401142Google ScholarDigital Library
- Shu Wu, Yuyuan Tang, Yanqiao Zhu, Liang Wang, Xing Xie, and Tieniu Tan. 2019a. Session-Based Recommendation with Graph Neural Networks. In AAAI. AAAI Press, 346--353. https://doi.org/10.1609/aaai.v33i01.3301346Google ScholarDigital Library
- Shu Wu, Yuyuan Tang, Yanqiao Zhu, Liang Wang, Xing Xie, and Tieniu Tan. 2019b. Session-Based Recommendation with Graph Neural Networks. In AAAI. AAAI Press, 346--353. https://doi.org/10.1609/aaai.v33i01.3301346Google ScholarDigital Library
- Mengqi Zhang, Shu Wu, Meng Gao, Xin Jiang, Ke Xu, and Liang Wang. 2022. Personalized Graph Neural Networks With Attention Mechanism for Session-Aware Recommendation. IEEE Trans. Knowl. Data Eng., Vol. 34, 8 (2022), 3946--3957. https://doi.org/10.1109/TKDE.2020.3031329Google ScholarCross Ref
Index Terms
- Optimizing Reciprocal Rank with Bayesian Average for improved Next Item Recommendation
Recommendations
Understanding item consumption orders for right-order next-item recommendation
Although the relevance problem in recommender systems, which typically refers to the similarity between the preference of the user and the items the system recommends, has been well studied, the issue of making recommendations in right orders has barely ...
Using a trust network to improve top-N recommendation
RecSys '09: Proceedings of the third ACM conference on Recommender systemsTop-N item recommendation is one of the important tasks of recommenders. Collaborative filtering is the most popular approach to building recommender systems which can predict ratings for a given user and item. Collaborative filtering can be extended ...
User Similarity Adjustment for Improved Recommendations
MIKE 2015: Proceedings of the Third International Conference on Mining Intelligence and Knowledge Exploration - Volume 9468Recommender systems are becoming more and more attractive in both research and commercial communities due to Information overload problem and the popularity of the Internet applications. Collaborative Filtering, a popular branch of recommendation ...
Comments