University of Limerick
Dolci_2024_Loss.pdf (878.77 kB)

Loss compensation in multi-session recommendation under limited item availability

Download (878.77 kB)
journal contribution
posted on 2024-06-26, 07:46 authored by Davide Azzalini, Fabio Azzalini, Chiara Criscuolo, Tommaso Dolci, Davide Martinenghi, Sihem Amer-Yahia

In many recommendation applications, items may have limited availability thereby causing conflict among users interested in the same items. Over time, this results in unequal user treatment: few users are recommended the limited items and receive preferential treatment, while the rest is left with sub-optimal recommendations, ultimately leading them to leave. In this paper, we formalize the novel problem of compensating users in multi-session recommendations under limited item availability. Our aim is to generate recommendations that not only optimize accuracy, but also compensate users over time for the loss of accuracy incurred in previous iterations. We design compensation strategies that serve users and items in different orders and accommodate various recommendation adoption models. Our al?gorithms are integrated into SoCRATe (System for Compensating Recommendations with Availability and Time), a framework that enables us to study loss compensation over time. Our experiments on real data demonstrate that to best compensate users for the incurred loss, traditional recommenders need to be revisited to account for item availability. Our experiments on synthetic data explore different parameters of our solution and show that it is much faster than an optimal (brute-force) compensation strategy, while achieving comparable results.



Proceedings of the 27th International Conference on Extending Database Technology (EDBT), 2024, pp. 522-533

Also affiliated with

  • LERO - The Science Foundation Ireland Research Centre for Software

Usage metrics

    University of Limerick


    Ref. manager