Kullanıcı Eğilimlerinin Göreceli Modellenmesinin Popülerlik Yanlılığına Etkisi

Öz

Dijitalleşmenin hızlanmasıyla kullanıcılar çok sayıda ürün ve hizmet seçeneğiyle karşı karşıya kalmakta, bu da kişiselleştirilmiş içeriklere erişimi zorlaştırmaktadır. Öneri sistemleri bu soruna çözüm sunmakla birlikte, geleneksel yaklaşımlar genellikle tek boyutlu puanlara dayanmakta ve popüler içerikleri öne çıkarma eğilimindedir. Bu durum önerilerin çeşitlilik ve adaletini sınırlandırmaktadır. Bu çalışmada, çok boyutlu veriler üzerinde kullanıcıların puanlama davranışlarını bağlamsal olarak analiz eden yeni bir yöntem olan RelPop önerilmektedir. RelPop, aynı puanın farklı kullanıcılar için göreli anlamını dikkate alarak içerikleri yeniden sıralamakta, böylece önerilerin özgünlük ve adalet düzeyi artmaktadır. Ayrıca, öneri listelerindeki popülerlik yanlılığını ölçmek için ADPI metriği geliştirilmiştir. İki çok ölçütlü veri seti üzerinde yapılan deneyler, RelPop’un önerilerin yenilik ve özgünlüğünü artırdığını, ADPI’nın ise popülerlik yanlılığını daha hassas biçimde ortaya koyduğunu göstermektedir.

Anahtar Kelimeler

• Çeşitlilik
• Çok ölçütlü öneri sistemleri
• Göreceli skor
• Popülerlik yanlılığı

The Effect of Relative Modeling of User Trends on Popularity Bias

Öz

Users are faced with an overwhelming number of products and services, making personalized content harder to access. Recommender systems address this challenge, but traditional approaches often rely on one-dimensional ratings and tend to prioritize popular items. This limits diversity and fairness in recommendations. To overcome these issues, this study proposes RelPop, a novel method that contextually analyzes users’ rating behaviors on multi-dimensional data. RelPop accounts for the relative meaning of the same score across different users, re-ranking items according to individual evaluation habits. This leads to more original, fair, and user-specific recommendations. Furthermore, a new performance metric, ADPI, is introduced to objectively measure popularity bias by evaluating how much recommended items deviate from the popularity center in a non-directional way. Experimental results on two multi-criteria datasets demonstrate that RelPop enhances novelty and uniqueness, while ADPI provides a more precise assessment of popularity bias.

Anahtar Kelimeler

• Diversity
• Multi-criteria recommender system
• Preference score
• Popularity bias

Kaynakça

1. 1. Zhang, S., Yao, L., Sun, A. & Tay, Y. (2019). Deep learning based recommender system: A survey and new perspectives. ACM Computing Surveys, 52(1), 1-38.
2. 2. Gomez-Uribe, C. A. & Hunt, N. (2016). The Netflix recommender system: Algorithms, business value, and innovation. ACM Transactions on Management Information Systems, 6(4), 1-19.
3. 3. Ricci, F., Rokach, L. & Shapira, B. (2022). Recommender systems: Techniques, applications, and challenges. In F. Ricci, L. Rokach & B. Shapira (Eds.), Recommender Systems Handbook (1-35). Cham: Springer.
4. 4. Schwartz, B. (2016). The paradox of choice: Why more is less (Rev. ed.). New York: Harper Perennial.
5. 5. Jannach, D. & Adomavicius, G. (2017). Price and profit awareness in recommender systems. In Proceedings of the 11th ACM Conference on Recommender Systems (192-200).
6. 6. Koren, Y. & Bell, R. (2015). Advances in collaborative filtering. In F. Ricci, L. Rokach & B. Shapira (Eds.), Recommender Systems Handbook (77-118). Cham: Springer.
7. 7. Adomavicius, G. & Kwon, Y. (2015). Multi-criteria recommender systems. In F. Ricci, L. Rokach & B. Shapira (Eds.), Recommender Systems Handbook (847-880). Cham: Springer.
8. 8. Baltrunas, L. & Ricci, F. (2014). Experimental evaluation of context-dependent collaborative filtering using item splitting. User Modeling and User-Adapted Interaction, 24(1-2), 7-34.

9. 9. Kalkan, İ.E. ve Şahin, C. (2023). Çapraz satışı destekleyebilecek transformer ile geliştirilmiş bir öneri sistemi. Çukurova Üniversitesi Mühendislik Fakültesi Dergisi, 38(2), 571-584.
10. 10. Liu, H., Wang, Y. & Chen, X. (2021). Multi-criteria recommendation with user preference learning. In Proceedings of the 15th ACM Conference on Recommender Systems (456-465).
11. 11. Jannach, D. & Adomavicius, G. (2016). Recommendations with a purpose. In Proceedings of the 10th ACM Conference on Recommender Systems (7-10).
12. 12. Koşar, O. ve Atak, M. (2023). Bulut bilişim sanal sunucu ürün seçiminde çok kriterli bir karar destek modeli. Çukurova Üniversitesi Mühendislik Fakültesi Dergisi, 38(4), 939-953.
13. 13. Baltrunas, L., Ludwig, B. & Ricci, F. (2011). Matrix factorization techniques for context-aware recommendation. In Proceedings of the 5th ACM Conference on Recommender Systems (301-304).
14. 14. Wang, X., He, X., Wang, M., Feng, F. & Chua, T. S. (2019). Neural graph collaborative filtering. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval (165-174).
15. 15. Steck, H. (2018). Calibrated recommendations. In Proceedings of the 12th ACM Conference on Recommender Systems (154-162).
16. 16. Ekstrand, M.D., Azpiazu, I.M., Anuyah, O., McNeill, D., Ekstrand, J.D. & Pera, M.S. (2018). All the cool kids, how do they fit in?: Popularity and demographic biases in recommender evaluation and effectiveness. In Proceedings of the Conference on Fairness, Accountability and Transparency (172-186).
17. 17. Abdollahpouri, H., Mansoury, M., Burke, R. & Mobasher, B. (2020). The connection between popularity bias, calibration, and fairness in recommendation. In Proceedings of the 14th ACM Conference on Recommender Systems (726-731).
18. 18. Klimashevskaia, A., Abdollahpouri, H. & Mobasher, B. (2022). Popularity bias in collaborative filtering: A survey. In Proceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization (1–11).
19. 19. Kowald, D. & Lacic, E. (2022). Popularity bias in different domains: A multi-domain analysis. In Proceedings of the 16th ACM Conference on Recommender Systems (1-11).
20. 20. Abdollahpouri, H., Burke, R. & Mobasher, B. (2019). The unfairness of popularity bias in recommendation. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society (148-154).
21. 21. Steck, H. (2011). Item popularity and recommendation accuracy. In Proceedings of the 5th ACM Conference on Recommender Systems (125-132).
22. 22. Wei, D., Wang, X., Li, Y. & He, X. (2021). Causal inference for recommender systems. In Proceedings of the 30th ACM International Conference on Information and Knowledge Management (3043-3049).
23. 23. Mansoury, M., Abdolllahpouri, H., Smith, J.,Dehpanah, A., Pechenizkiy, M., Mobasher, B. (2019). Investigating potential factors associated with gender discrimination in collaborative recommender systems. Information Processing & Management, 57(2), 102371.
24. 24. Taçlı, Y., Yalçın, E. & Bilge, A. (2022). Novel approaches to measuring the popularity inclination of users for the popularity bias problem. In Proceedings of the 6th International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT) (555-560).
25. 25. Lee, J., Lee, D., Lee, Y., Hwang, W. & Kim, S. (2016). Improving the accuracy of top-n recommendation using a preference model. Information Sciences, 348, 290-304.
26. 26. Yalcin, E. & Bilge, A. (2021). Investigating and counteracting popularity bias in group recommendations. Information Processing & Management, 58(5), 102608.
27. 27. Zhang, Y., Chen, X., Ai, Q., Croft, W. B. & Guo, J. (2017). Towards conversational search and recommendation: System ask, user respond. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management (177-186).
28. 28. Abdollahpouri, H., Burke, R. & Mobasher, B. (2017). Controlling popularity bias in learning-to-rank recommendation. In Proceedings of the 11th ACM Conference on Recommender Systems (RecSys) (42-46).
29. 29. Lakiotaki, K., Matsatsinis, N.F. & Tsoukias, A. (2011). Multicriteria user modeling in recommender systems. IEEE Intelligent Systems, 26(2), 64-76.
30. 30. Wang, H., Lu, Y. & Zhai, C. (2011). Latent aspect rating analysis without aspect keyword supervision. In Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (618-626).
31. 31. Herlocker, J.L., Konstan, J.A., Terveen, L.G. & Riedl, J.T. (2004). Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems (TOIS), 22(1), 5-53.
32. 32. Bellogín, A., Castells, P. & Cantador, I. (2011). Precision-oriented evaluation of recommender systems: An algorithmic comparison. In Proceedings of the 5th ACM Conference on Recommender Systems (333-336).
33. 33. Vargas, S. & Castells, P. (2011). Rank and relevance in novelty and diversity metrics for recommender systems. In Proceedings of the 5th ACM Conference on Recommender Systems (109-116).
34. 34. Park, Y.J. & Tuzhilin, A. (2008). The long tail of recommender systems and how to leverage it. In Proceedings of the 2008 ACM Conference on Recommender Systems (11-18).
35. 35. Burke, R. (2002). Hybrid recommender systems: Survey and experiments. User Modeling and User-Adapted Interaction, 12(4), 331-370.
36. 36. Breiman, L., Friedman, J., Stone, C.J. & Olshen, R.A. (1984). Classification and regression trees. Chapman & Hall/CRC.