Yao Wu
Skip Abstract Section
Abstract
When only users’ preferences and interests are considered by a recommendation algorithm, it will lead to the severe long-tail problem over items. Therefore, the unfair exposure phenomenon of recommended items caused by this problem has attracted widespread attention in recent years. For the first time, we reveal the fact that there is a more serious unfair exposure problem in session-based recommender systems (SRSs), which learn the short-term and dynamic preferences of users from anonymous sessions. Considering the fact that in SRSs, recommendations are provided multiple times and item exposures are accumulated over interactions in a session, we define new metrics both for the fairness of item exposure and recommendation quality among sessions. Moreover, we design a dynamic Fairness-Assurance STrategy for sEssion-based Recommender systems (FASTER). FASTER is a post-processing strategy that tries to keep a balance between item exposure fairness and recommendation quality. It can also maintain the fairness of recommendation quality among sessions. The effectiveness of FASTER is verified on three real-world datasets and five original algorithms. The experiment results show that FASTER can generally reduce the unfair exposure of different session-based recommendation algorithms while still ensuring a high level of recommendation quality.
- [1] . 2018. A reductions approach to fair classification. In Proceedings of the International Conference on Machine Learning. PMLR, 60–69.Google Scholar
- [2] . 2019. Designing fair ranking schemes. In Proceedings of the International Conference on Management of Data. ACM, 1259–1276.Google ScholarDigital Library
- [3] . 2011. The form of the forgetting curve and the fate of memories. J. Math. Psychol. 55, 1 (2011), 25–35.Google ScholarCross Ref
- [4] . 2019. Fairness in recommendation ranking through pairwise comparisons. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2212–2220.Google ScholarDigital Library
- [5] . 2018. Equity of attention: Amortizing individual fairness in rankings. In Proceedings of the 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. ACM, 405–414.Google ScholarDigital Library
- [6] . 2021. DeepFair: Deep learning for improving fairness in recommender systems. Int. J. Interact. Multim. Artif. Intell. 6, 6 (2021), 86–95.Google Scholar
- [7] . 2021. Connecting user and item perspectives in popularity debiasing for collaborative recommendation. Inf. Process. Manag. 58, 1 (2021), 102387.Google ScholarCross Ref
- [8] . 2019. Compositional fairness constraints for graph embeddings. In Proceedings of the International Conference on Machine Learning. PMLR, 715–724.Google Scholar
- [9] . 2017. Multisided fairness for recommendation. arXiv preprint arXiv:1707.00093 (2017).Google Scholar
- [10] . 2017. Optimized pre-processing for discrimination prevention. In Proceedings of the International Conference on Advances in Neural Information Processing Systems. 3992–4001.Google Scholar
- [11] . 2017. Fair sharing for sharing economy platforms. In Fairness, Accountability and Transparency in Recommender Systems-Workshop on Responsible Recommendation.Google Scholar
- [12] . 2023. Bias and debias in recommender system: A survey and future directions. ACM Trans. Inf. Syst. 41, 3 (2023), 1–39.Google ScholarDigital Library
- [13] . 2022. S-Walk: Accurate and scalable session-based recommendation with random walks. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining. 150–160.Google ScholarDigital Library
- [14] . 2013. Biased assimilation, homophily, and the dynamics of polarization. Proc. Nat. Acad. Sci. 110, 15 (2013), 5791–5796.Google ScholarCross Ref
- [15] . 2021. Towards long-term fairness in recommendation. In Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 445–453.Google ScholarDigital Library
- [16] . 2019. Fairness-aware ranking in search & recommendation systems with application to LinkedIn talent search. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2221–2231.Google ScholarDigital Library
- [17] . 2020. Predicting session length for product search on e-commerce platform. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 1713–1716.Google ScholarDigital Library
- [18] . 1976. Repetition and memory. Psychol. Learn. Motivat. 10 (1976), 47–91.Google ScholarCross Ref
- [19] . 2007. Evaluating the accuracy of implicit feedback from clicks and query reformulations in web search. ACM Trans. Inf. Syst. 25, 2 (2007), 7.Google ScholarDigital Library
- [20] . 2007. Search engines that learn from implicit feedback. Computer 40, 8 (2007), 34–40.Google ScholarDigital Library
- [21] . 2018. Using image fairness representations in diversity-based re-ranking for recommendations. In Adjunct Publication of the 26th Conference on User Modeling, Adaptation and Personalization. ACM, 23–28.Google ScholarDigital Library
- [22] . 2015. Unequal representation and gender stereotypes in image search results for occupations. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. 3819–3828.Google ScholarDigital Library
- [23] . 2022. An attribute-driven mirror graph network for session-based recommendation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1674–1683.Google ScholarDigital Library
- [24] . 2020. Multi-objective optimization for guaranteed delivery in video service platform. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 3017–3025.Google ScholarDigital Library
- [25] . 2022. FairSR: Fairness-aware sequential recommendation through multi-task learning with preference graph embeddings. ACM Trans. Intell. Syst. Technol. 13, 1 (2022), 1–21.Google ScholarDigital Library
- [26] . 2017. Neural attentive session-based recommendation. In Proceedings of the ACM Conference on Information and Knowledge Management. 1419–1428.Google ScholarDigital Library
- [27] . 2003. Amazon.com recommendations: Item-to-item collaborative filtering. IEEE Internet Comput. 7, 1 (2003), 76–80.Google ScholarDigital Library
- [28] . 2018. STAMP: Short-term attention/memory priority model for session-based recommendation. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 1831–1839.Google ScholarDigital Library
- [29] . 2018. Personalizing fairness-aware re-ranking. arXiv preprint arXiv:1809.02921 (2018).Google Scholar
- [30] . 2020. Controlling fairness and bias in dynamic learning-to-rank. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 429–438.Google ScholarDigital Library
- [31] . 2015. Replication and analysis of Ebbinghaus’ forgetting curve. PloS One 10, 7 (2015), e0120644.Google ScholarCross Ref
- [32] . 2020. Correcting for selection bias in learning-to-rank systems. In Proceedings of the Web Conference. 1863–1873.Google ScholarDigital Library
- [33] . 2022. Heterogeneous global graph neural networks for personalized session-based recommendation. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining. 775–783.Google ScholarDigital Library
- [34] . 2020. FairRec: Two-sided fairness for personalized recommendations in two-sided platforms. In Proceedings of the Web Conference. 1194–1204.Google ScholarDigital Library
- [35] . 2019. Rethinking the item order in session-based recommendation with graph neural networks. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 579–588.Google ScholarDigital Library
- [36] . 2022. Fair-SRS: A fair session-based recommendation system. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining. 1601–1604.Google Scholar
- [37] . 2019. Fighting fire with fire: Using antidote data to improve polarization and fairness of recommender systems. In Proceedings of the 12th ACM International Conference on Web Search and Data Mining. ACM, 231–239.Google ScholarDigital Library
- [38] . 2009. BPR: Bayesian personalized ranking from implicit feedback. In Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence. 452–461.Google ScholarDigital Library
- [39] . 2010. Factorizing personalized Markov chains for next-basket recommendation. In Proceedings of the 19th International Conference on World Wide Web. 811–820.Google ScholarDigital Library
- [40] . 2005. An MDP-based recommender system. J. Mach. Learn. Res. 6, 9 (2005).Google Scholar
- [41] . 2020. Opportunistic multi-aspect fairness through personalized re-ranking. In Proceedings of the 28th ACM Conference on User Modeling, Adaptation and Personalization. 239–247.Google ScholarDigital Library
- [42] . 1974. The use of field experimental observations in estimating advertising recall. J. Market. Res. 11, 4 (1974), 369–378.Google ScholarCross Ref
- [43] . 2017. Predicting session length in media streaming. In Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval. 977–980.Google ScholarDigital Library
- [44] . 2021. A survey on session-based recommender systems. ACM Comput. Surv. 54, 7 (2021), 1–38.Google ScholarDigital Library
- [45] . 2019. Modeling multi-purpose sessions for next-item recommendations via mixture-channel purpose routing networks. In Proceedings of the International Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence.Google ScholarCross Ref
- [46] . 2023. A survey on the fairness of recommender systems. ACM J. ACM 41, 3 (2023), 1–43.Google Scholar
- [47] . 2020. Global context enhanced graph neural networks for session-based recommendation. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 169–178.Google ScholarDigital Library
- [48] . 2021. Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 1791–1800.Google ScholarDigital Library
- [49] . 2021. Fairness-aware news recommendation with decomposed adversarial learning. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35. 4462–4469.Google ScholarCross Ref
- [50] . 2017. Session-aware information embedding for e-commerce product recommendation. In Proceedings of the ACM on Conference on Information and Knowledge Management. 2379–2382.Google ScholarDigital Library
- [51] . 2019. Session-based recommendation with graph neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. 346–353.Google ScholarDigital Library
- [52] . 2020. FAST: A fairness assured service recommendation strategy considering service capacity constraint. In Proceedings of the International Conference on Service-oriented Computing. Springer, 287–303.Google ScholarDigital Library
- [53] . 2021. Self-supervised graph co-training for session-based recommendation. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2180–2190.Google ScholarDigital Library
- [54] . 2017. Fairness beyond disparate treatment & disparate impact: Learning classification without disparate mistreatment. In Proceedings of the 26th International Conference on World Wide Web. International World Wide Web Conferences Steering Committee, 1171–1180.Google ScholarDigital Library
- [55] . 2020. Reducing disparate exposure in ranking: A learning to rank approach. In Proceedings of the Web Conference. 2849–2855.Google ScholarDigital Library
- [56] . 2019. Feature-level deeper self-attention network for sequential recommendation. In Proceedings of the International Joint Conference on Artificial Intelligence. 4320–4326.Google ScholarCross Ref
- [57] . 2018. FMSR: A fairness-aware mobile service recommendation method. In Proceedings of the IEEE International Conference on Web Services (ICWS). IEEE, 171–178.Google ScholarCross Ref
- [58] . 1959. The remembering and forgetting of advertising. J. Market. 23, 3 (1959), 239–243.Google ScholarCross Ref
- [59] . 1980. Remembering and forgetting television ads. J. Advert. Res. 20, 2 (1980), 7–13.Google Scholar
Index Terms
- FASTER: A Dynamic Fairness-assurance Strategy for Session-based Recommender Systems
Recommendations
Diversification in session-based news recommender systems
AbstractRecommender systems are widely applied in digital platforms such as news websites to personalize services based on user preferences. In news websites, most of users are anonymous and the only available data is sequences of items in anonymous ...
Investigating serendipity in recommender systems based on real user feedback
SAC '18: Proceedings of the 33rd Annual ACM Symposium on Applied ComputingOver the past several years, research in recommender systems has emphasized the importance of serendipity, but there is still no consensus on the definition of this concept and whether serendipitous items should be recommended is still not a well-...
A fairness-aware multi-stakeholder recommender system
AbstractTraditional recommender systems mainly focus on the accuracy of recommendation, which lead to recommender systems reinforcing popular items and ignoring lesser-known items. There is increasing evidence that providing good recommendations of ...
Comments