Green Recommender Systems: Optimizing Dataset Size for Energy-Efficient Algorithm Performance

TL;DR

This paper explores how reducing dataset sizes through downsampling can maintain recommendation quality while significantly decreasing energy consumption in recommender systems.

Contribution

It demonstrates that strategic dataset reduction can preserve recommendation performance and improve energy efficiency in green recommender systems.

Findings

FunkSVD and BiasedMF maintain high performance with up to 50% data reduction.

Dataset downsampling can reduce training costs by over 50%.

Recommendation quality remains within 13% of full dataset performance with reduced data.

Abstract

As recommender systems become increasingly prevalent, the environmental impact and energy efficiency of training large-scale models have come under scrutiny. This paper investigates the potential for energy-efficient algorithm performance by optimizing dataset sizes through downsampling techniques in the context of Green Recommender Systems. We conducted experiments on the MovieLens 100K, 1M, 10M, and Amazon Toys and Games datasets, analyzing the performance of various recommender algorithms under different portions of dataset size. Our results indicate that while more training data generally leads to higher algorithm performance, certain algorithms, such as FunkSVD and BiasedMF, particularly with unbalanced and sparse datasets like Amazon Toys and Games, maintain high-quality recommendations with up to a 50% reduction in training data, achieving nDCG@10 scores within approximately 13% of full dataset performance. These findings suggest that strategic dataset reduction can decrease computational and environmental costs without substantially compromising recommendation quality. This study advances sustainable and green recommender systems by providing insights for reducing energy consumption while maintaining effectiveness.