Balancing carbon footprint and algorithm performance in recommender systems: A comprehensive benchmark

In this paper, we present a reproducible pipeline to benchmark the trade-off between carbon emissions and recommendation performance across 14 algorithms and three publicly available datasets. In particular, we contribute: (a) a standardized protocol to account for carbon emissions of recommendation algorithms; (b) an empirical quantification of the carbon cost of hyperparameter tuning, and (c) an evaluation of data-reduction strategies as a low-cost approach to reduce emissions while improving certain non-accuracy metrics. Unlike previous literature, which mainly focused on the trade-off between performance and emissions, our benchmark reveals the cost of hyperparameter tuning. It examines the impact of data reduction techniques on the path toward sustainability-aware recommender systems. Our results show that simpler algorithms often deliver competitive accuracy at significantly lower emissions, and that exhaustive tuning can dramatically increase carbon costs with limited accuracy gains. Generally speaking, this study aims to discuss the challenges of energy consumption in recommender systems and to develop a new generation of algorithms that prioritize sustainability. All code and experiment traces are publicly released for reproducibility on Github. 1 1 https://github.com/swapUniba/RecSysCarbonFootprint . • Benchmark of 14 algorithms analyzing the trade-off between accuracy and environmental impact across multiple datasets. • Quantification of carbon emissions associated with tuning and training, showing that hyperparameter tuning can drastically increase emissions with limited performance gains. • Exploration of lightweight strategies, such as data reduction, to develop more sustainable and diverse recommender systems.