A Multi-Criteria Recommendation Systems Based on Behavioral Tendencies: BADPM Model

Abstract

Although contemporary recommender systems achieve highly effective recommendation performance, they continue to suffer from popularity bias—a pervasive issue that restricts users’ exposure to less popular yet potentially relevant items. This study develops a mechanism that addresses popularity bias in multi-criteria recommendation systems, increasing satisfaction and including unpopular products in the list. The proposed method is based on the tendency for individuals to prioritize the preferences of users who are similar to their own tastes and preferences in their decision-making processes, while only slightly considering the evaluations of users with opposing tendencies. Furthermore, it is assumed that products negatively evaluated by users with opposing tendencies can sometimes become more attractive to the target user. Building on this behavioral foundation, the approach considers both the preferences of similar users and the evaluations of users with opposing tendencies when generating a recommendation list. Comparative analyses of the proposed method with traditional methods are conducted on three different datasets, and the results demonstrate that the proposed approach is successful in reducing popularity bias and improving recommendation quality.

Keywords

• Multi-criteria recommender systems
• Popularity bias
• Personalized recommendations
• Behavioral modeling

Davranışsal Eğilimlere Dayalı Çok Ölçütlü Öneri Sistemleri: BADPM Modeli

Abstract

Günümüzde öneri sistemleri oldukça başarılı tavsiyeler üretse de kullanıcıların daha az bilinen ancak potansiyel olarak ilgi çekici ürünlere erişimini kısıtlayan önemli bir sorun olan popülerlik yanlılığı ile karşı karşıya kalmaktadır. Bu çalışmada çok ölçütlü öneri sistemlerinde popülerlik yanlılığı probleminin ele alındığı, memnuniyetin arttırılıp buna ek olarak listede popüler olmayan ürünlerin yer aldığı bir mekanizma geliştirilmiştir. Önerilen yöntemde, bireylerin karar verme süreçlerinde kendi zevk ve eğilimlerine benzeyen kullanıcıların tercihlerine daha fazla önem vermesine, buna karşılık zıt eğilimlere sahip kullanıcıların değerlendirmelerini sınırlı düzeyde dikkate alma eğilimine dayanmaktadır. Dahası, zıt eğilimli kullanıcıların olumsuz değerlendirdiği ürünlerin kimi durumlarda hedef kullanıcı için daha çekici hâle gelebileceği varsayılmaktadır. Bu davranışsal temelden hareketle geliştirilen yaklaşım, öneri listesi oluştururken hem benzer kullanıcıların tercihlerini hem de zıt eğilimlere sahip kullanıcıların değerlendirmelerini birlikte ele almaktadır. Üç farklı veri seti üzerinde önerilen yöntem ile klasik yöntemlerin karşılaştırmalı analiz sonuçları yapılmış ve elde edilen sonuçlar, önerilen yaklaşımın popülerlik yanlılığını azaltmada ve öneri kalitesini artırmada başarılı olduğunu göstermiştir.

Keywords

• Çok ölçütlü öneri sistemleri
• Popülerlik yanlılığı
• Kişiselleştirilmiş öneriler
• Davranışsal modelleme

References

1. Abdollahpouri, H., Burke, R., & Mobasher, B. (2017). Controlling popularity bias in learning-to-rank recommendation. RecSys 2017: Proceedings of the 11th ACM Conference on Recommender Systems, 42–46.
2. Abdollahpouri, H., Burke, R., & Mobasher, B. (2018). Popularity-aware item weighting for long-tail recommendation. arXiv. https://arxiv.org/abs/1802.05382.
3. Abdollahpouri, H., Burke, R., & Mobasher, B. (2019). Managing popularity bias in recommender systems with personalized re-ranking. arXiv preprint arXiv:1901.07555.
4. Abdollahpouri, H., Mansoury, M., Burke, R., & Mobasher, B. (2019b). The unfairness of popularity bias in recommendation. arXiv preprint arXiv:1907.13286.
5. Abdollahpouri, H., Mansoury, M., Burke, R., Mobasher, B., & Malthouse, E. (2021). User-centered evaluation of popularity bias in recommender systems. In Proceedings of the 29th ACM conference on user modeling, adaptation and personalization (pp. 119-129).
6. Balli, N., & Kaya, T. T. (2025). Kullanıcı eğilimlerinin göreceli modellenmesinin popülerlik yanlılığına etkisi. Cukurova University Faculty of Engineering Journal, 40(3), 671–686.
7. Cremonesi, P., Koren, Y., & Turrin, R. (2010). Performance of recommender algorithms on top-n recommendation tasks. In Proceedings of the fourth ACM conference on Recommender systems (pp. 39-46).
8. Isinkaye, F. O., Folajimi, Y., & Ojokoh, B. (2015). Recommendation systems: Principles, methods and evaluation. Egyptian Informatics Journal, 16, 261–273.

9. Jannach, D., Karakaya, Z., & Gedikli, F. (2012). Accuracy improvements for multi-criteria recommender systems. Proceedings of the 13th ACM Conference on Electronic Commerce, 674–689.
10. Kang, Z., Peng, C., Yang, M., & Cheng, Q. (2016). Top-n recommendation on graphs. Proceedings of the 25th ACM CIKM, 2101–2106.
11. Kaya, T. T. (2025a). Mitigating popularity bias in fair and explainable recommender systems using SHAP. Duzce University Journal of Science and Technology, 13(3), 1180–1199.
12. Kaya, T. T. (2025b). A novel clustering-based multi-criteria group recommender system. Neurocomputing, 132329.
13. Kaya, T. T. (2025c). Analysis of Performance and Popularity Bias in Recommender Systems Based on Personality Traits. In SETSCI-Conference Proceedings (Vol. 22, pp. 11-15). SETSCI-Conference Proceedings.
14. 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.
15. Pichl, M., Zangerle, E., & Specht, G. (2014). Combining Spotify and Twitter Data for Generating a Recent and Public Dataset for Music Recommendation. In Grundlagen von Datenbanken (pp. 35-40).
16. Ren, Y., Li, G., & Zhou, W. (2012). Learning user preference patterns for Top-N recommendations. In 2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology (Vol. 1, pp. 137-144). IEEE.
17. Sanders, R. (1987). The Pareto principle: Its use and abuse. Journal of Services Marketing, 1(2), 37–40.
18. Tacli, Y., Yalcin, E., & Bilge, A. (2022). Novel approaches to measuring the popularity inclination of users for the popularity bias problem. In 2022 International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT) (pp. 555-560). IEEE.
19. Ünver, Ö., & Gamgam, H. (2008). Uygulamalı temel istatistik yöntemler.
20. 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 (pp. 618-626).
21. Xue, F., He, X., Wang, X., Xu, J., Liu, K., & Hong, R. (2019). Deep item-based collaborative filtering for top-n recommendation. ACM Transactions on Information Systems (TOIS), 37(3), 1-25.
22. Yalcin, E. (2021). Blockbuster: A new perspective on popularity-bias in recommender systems. In 2021 6th International Conference on Computer Science and Engineering (UBMK) (pp. 107-112). IEEE.
23. Yalcin, E., & Bilge, A. (2022). Evaluating unfairness of popularity bias in recommender systems: A comprehensive user-centric analysis. Information Processing & Management, 59(6), 103100.
24. Yalcin, E., & Bilge, A. (2023). Popularity bias in personality perspective: An analysis of how personality traits expose individuals to the unfair recommendation. Concurrency and Computation: Practice and Experience, 35(9), e7647.