Learning data characteristics of a session-based recommendation system and their impact on system performance

• Autorzy: Kużelewska Urszula

Identyfikatory

DOI

10.24425/ijet.2025.153579

• Języki publikacji: EN

Abstrakty

EN

Recommendation systems are the most effective solution for enhancing user satisfaction and personalising e-commerce services on the internet. These systems use advanced procedures to analyse massive volumes of data, ensuring users receive the most relevant and suitable products available. The success of recommendation systems hinges on the quality of the methods used. However, there is also an impact on the input data. Session-based techniques are the most effective way to generate recommendations. They focus on short-term user interactions organised in sessions. This procedure is the best for real-world scenarios, where one-time users and limited item availability are prevalent. The objective of this study is to examine the relationship between data metrics, including density, shape, and popularity, and the performance of session-based algorithms, in terms of accuracy and coverage.

Słowa kluczowe

EN

session-based recommenders; evaluation of recommendations; data metrics

• Wydawca: Polish Academy of Sciences, Committee of Electronics and Telecommunication
• Czasopismo: International Journal of Electronics and Telecommunications
• Rocznik: 2025
• Tom: Vol. 71, No. 2
• Strony: 345--352
• Opis fizyczny: Bibliogr. 28 poz., tab.

Twórcy

autor

Kużelewska Urszula

• Bialystok University of Technology, Poland

Bibliografia

• [1] D. Jannach, M. Zanker, A. Felfernig, and G. Friedrich, Recommender Systems. Cambridge University Press, 2010, vol. 24.
• [2] F. Ricci, L. Rokach, and B. Shapira, Recommender Systems: Introduction and Challenges. Springer, 2015, vol. 1-35.
• [3] J. B. Schafer, D. Frankowski, J. Herlocker, and S. Sen, Collaborative Filtering Recommender Systems. Springer Berlin Heidelberg, 2007.
• [4] D. Jannach, B. Mobasher, and S. Berkovsky, “Research directions in session-based and sequential recommendation,” User Model.-User-Adap. Interaction, vol. 30, p. 609-616, 2020.
• [5] S. Ozdemir and D. Susarla, Feature Engineering Made Easy: Identify unique features from your dataset in order to build powerful machine learning systems. Packt Publishing Ltd, 2018.
• [6] G. Shani and A. Gunawardana, Evaluating Recommendation Systems, 2011, vol. 12, pp. 257-297.
• [7] R. Ayub, M. a. Ghazanfar, Z. Mehmood, T. Saba, R. Alharbey, A. Munshi, and M. Alrige, “Modeling user rating preference behavior to improve the performance of the collaborative filtering based recommender systems,” PLOS ONE, vol. 14, 08 2019.
• [8] G. Adomavicius and J. Zhang, “Impact of data characteristics on recommender systems performance,” ACM Trans. Manage. Inf. Syst., vol. 3, no. 1, 2012.
• [9] Y. Deldjoo, A. Bellogin, and T. Di Noia, “Explaining recommender systems fairness and accuracy through the lens of data characteristics,” Information Processing & Management, vol. 58, no. 5, p. 102662, 2021.
• [10] “Diginetica.” [Online]. Available: https://darel13712.github.io/rs datasets/Datasets/diginetica/
• [11] U. Ku˙zelewska and M. Charytanowicz, “Characteristics of the learning data of a session-based recommendation system and their impact on the performance of the system,” in Proceeding of 32nd International Conference on Information Systems Development, 09 2024.
• [12] Y. Deldjoo, T. Di Noia, E. Di Sciascio, and F. A. Merra, “How dataset characteristics affect the robustness of collaborative recommendation models,” in Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, ser. SIGIR’20. Association for Computing Machinery, 2020, p. 951-960.
• [13] C.-N. Hsu, H.-H. Chung, and H.-S. Huang, “Mining skewed and sparse transaction data for personalized shopping recommendation,” Machine Learningk, vol. 57, pp. 35-59, 01 2004.
• [14] S. Shaikh, V. R. Kagita, V. Kumar, and A. K. Pujari, “Data augmentation and refinement for recommender system: A semi-supervised approach using maximum margin matrix factorization,” Expert Systems with Applications, vol. 238, p. 121967, 2024.
• [15] W. Yang, S. Fan, and H. Wang, “An item-diversity-based collaborative filtering algorithm to improve the accuracy of recommender system,” in 2018 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 2018, pp. 106-110.
• [16] J. Y. Chin, Y. Chen, and G. Cong, “The datasets dilemma: How much do we really know about recommendation datasets?” ser. WSDM ’22. Association for Computing Machinery, 2022, p. 141-149.
• [17] M. Ludewig, N. Mauro, S. Latifi, and D. Jannach, “Empirical analysis of session-based recommendation algorithms,” User Model User-Adap Inter, vol. 31, p. 149-181, 2021.
• [18] K. Verstrepen and B. Goethals, “Unifying nearest neighbors collaborative filtering,” in Proceedings of the 8th ACM Conference on Recommender Systems, ser. RecSys ’14. Association for Computing Machinery, 2014, p. 177-184.
• [19] M. Ludewig and D. Jannach, “Evaluation of session-based recommendation algorithms,” User Modeling and User-Adapted Interaction, vol. 28, no. 4-5, p. 331-390, 2018.
• [20] B. Hidasi, A. Karatzoglou, L. Baltrunas, and D. Tikk, “Session-based recommendations with recurrent neural networks,” in Proceedings International Conference on Learning Representations, ser. ICLR ’16, 2016.
• [21] C. C. Aggarwal, Recommender Systems. The Textbook. Springer, 2016.
• [22] K. Cho, B. Merrienboer, C. Gulcehre, F. Bougares, H. Schwenk, and Y. Bengio, “Learning phrase representations using rnn encoder-decoder for statistical machine translation,” p. 1724-1734, 2014.
• [23] Q. Liu, Y. Zeng, R. Mokhosi, and H. Zhang, “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, ser. KDD ’18. Association for Computing Machinery, 2018, p. 1831-1839.
• [24] K. J¨arvelin and J. Kek¨al¨ainen, “Cumulated gain-based evaluation of ir techniques,” ACM Trans. Inf. Syst., vol. 20, no. 4, p. 422-446, 2002.
• [25] G. Adomavicius and Y. Kwon, “Improving aggregate recommendation diversity using ranking-based techniques,” IEEE Transactions on Knowledge and Data Engineering, vol. 24, no. 5, pp. 896-911, 2012.
• [26] M. Quadrana, P. Cremonesi, and D. Jannach, “Sequence-aware recommender systems,” ACM Comput. Surv., vol. 51, no. 4, 2018.
• [27] Smyth, Barry and McClave, Paul, “Similarity vs. diversity,” in Proceedings of the International Conference on Case-Based Reasoning. Springer, 2001, pp. 347-361.
• [28] “Session-rec software.” [Online]. Available: https://github.com/rn5l/session-rec

Uwagi

1. Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).

2. The work was supported by a grant from the Bialystok University of Technology WZ/WI-IIT/3/2023 and funded with resources for research by the Ministry of Education and Science in Poland.