Journal of Information Technology Applications and Management

  • 제30권4호
  • /
  • Pages.29-38
  • /
  • 2023
  • /
  • 1598-6284(pISSN)
  • /
  • 2508-1209(eISSN)
한국데이터전략학회 (Korea Data Strategy Society)
권호 표지
DOI QR코드

DOI QR Code

반사실적 데이터 증강에 기반한 인과추천모델: CausRec사례

A Causal Recommendation Model based on the Counterfactual Data Augmentation: Case of CausRec

  • 투고 : 2023.06.19
  • 심사 : 2023.08.03
  • 발행 : 2023.08.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

A single-learner model which integrates the user's positive and negative perceptions is proposed by augmenting counterfactual data to the interaction data between users and items, which are mainly used in collaborative filtering in this study. The proposed CausRec showed superior performance compared to the existing NCF model in terms of F1 value and AUC in experiments using three published datasets: MovieLens 100K, Amazon Gift Card, and Amazon Magazine. Compared to the existing NCF model, the F1 and AUC values of CausRec showed 1.2% and 2.6% performance improvement in MovieLens 100K data, and 2.2% and 10% improvement in Amazon Gift Card data, respectively. In particular, in experiments using Amazon Magazine data, F1 and AUC values were improved by 11.7% and 21.9%, respectively, showing a significant performance improvement effect. The performance of CausRec is improved because both positive and negative perceptions of the item were reflected in the recommendation at the same time. It is judged that the proposed method was able to improve the performance of the collaborative filtering because it can simultaneously alleviate the sparsity and imbalance problems of the interaction data.

키워드

과제정보

This work was supported by 2022 Hannam University Research Fund.

참고문헌

  1. Bonner, S. and Vasile, F., "Causal embeddings for recommendation", In RecSys, 2018, pp. 104-112.
  2. Chen, J., Dong, H., Qiu, Y., He, X., Xin, X., Chen, L., Lin, G., and Yang, K., "Autodebias: Learning to debias for recommendation", In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2021, pp. 21-30.
  3. Chen, L., Yan, X., Xiao, J., Zhang, H., Pu, S., and Zhuang, Y., "Counterfactual Samples Synthesizing for Robust Visual Question Answering", In CVPR, 2020.
  4. Herlocker, J., Konstan, J., Borchers, A., and Riedl, J.. "An Algorithmic Framework for Performing Collaborative Filtering", Proceedings of the 1999 Conference on Research and Development in Information Retrieval, 1999.
  5. Joachims, T., Swaminathan, A., and Schnabel, T., "Unbiased learning-to-rank with biased feedback", In WSDM, 2017, pp. 781-789.
  6. Rosenfeld, N., Mansour, Y., and Yom-Tov, E., "Predicting counterfactuals from large historical data and small randomized trials", in Proceedings of the 26th International Conference on World Wide Web Companion, 2017, pp. 602-609.
  7. Sato, M., Singh, J., Takemori, S., Sonoda, T., Zhang, Q., and Ohkuma, T., "Uplift-based evaluation and optimization of recommenders", In RecSys, 2019, pp. 296-304.
  8. Saito, Y., Yaginuma, S., Nishino, Y., Sakata, H., and Nakata, K., "Unbiased recommender learning from missing-not-at-random implicit feedback", In Proceedings of the 13th International Conference on Web Search and Data Mining, 2020, pp. 501-509.
  9. Wang, X., Bendersky, M., Metzler, D., and Najork, M., "Learning to rank with selection bias in personal search", In SIGIR, 2016, pp. 115-124.
  10. Wang, Y., Qin, Y., Han, Y., Yin, M., Zhou, J., Yang, H., and Zhang, M., "AD-AUG: Adversarial Data Augmentation for Counterfactual Recommendation", In ECML PKDD, 2023.
  11. Wang, Z., Zhang, Z., Xu, H., Chen, X, Zhang, Y., Zhao, W. X., and Wen, J., "Counterfactual Data-Augmented Sequential Recommendation", In SIGIR, 2021, pp. 11-15.
  12. Xie, X., Liu, Z., Wu, S., Sun, F., Liu, C., Chen, J., Gao, J., Cui, B., and Ding, B.. "CausCF: Causal Collaborative Filtering for Recommendation Effect Estimation", In Proceedings of the 30th ACM International Conference on Information & Knowledge Management (CIKM '21), 2021, pp. 4253-4263.
  13. Xu, S., Ge, Y., Li, Y., Fu, Z., Chen, X., and Zhang, Y., "Causal collaborative filtering", arXiv preprint arXiv:2102.01868, 2021.
  14. Yuan, B., Hsia, J., Yang, M., Zhu, H., Chang, C., Dong, Z., and Lin, C., "Improving ad click prediction by considering non-displayed events", In Proceedings of the 28th ACM International Conference on Information and Knowledge Management, 2019, pp. 329-338.
  15. Zhang, Y., Feng, F., He, X., Wei, T., Song, C., Ling, G., and Zhang, Y., "Causal intervention for leveraging popularity bias in recommendation", In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2021, pp. 11-20.