14 Kg of CO2: Analyzing the Carbon Footprint and Performance of Session-Based Recommendation Algorithms

dc.catalogadorpva
dc.contributor.authorPlaza Carrasco, Alejandro
dc.contributor.authorGil, Juan Carlos
dc.contributor.authorParra Santander, Denis
dc.date.accessioned2025-04-28T16:39:13Z
dc.date.available2025-04-28T16:39:13Z
dc.date.issued2025
dc.description.abstractGreen AI aims to develop accurate AI models that are also sustainable without compromising the environment, especially in terms of carbon emissions. There are few studies on this topic in recommender systems, so we analyzed the trade-offs between recommendation performance and carbon footprint in session-based recommender systems. We use five public e-commerce datasets to predict the next item a user will interact with based solely on their past click events. The GRU4Rec algorithm and five unofficial reimplementations in different deep learning frameworks (Theano, PyTorch, TensorFlow, Keras, and Reckpack) are evaluated. The results indicate a strong effect of the loss function and dataset size on the carbon footprint without significantly affecting the accuracy metrics. We show evidence that the implementation choice for the same algorithm strongly affects the CO2 emitted, and optimized implementations do not sacrifice recommendation efficiency, which should be considered when choosing a framework or implementation for an algorithm.
dc.format.extent12 páginas
dc.fuente.origenORCID
dc.identifier.doi10.1007/978-3-031-87654-7_12
dc.identifier.eisbn978-3-031-87654-7
dc.identifier.isbn978-3-031-87653-0
dc.identifier.urihttps://doi.org/10.1007/978-3-031-87654-7_12
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/103484
dc.information.autorucEscuela de Ingeniería; Plaza Carrasco, Alejandro; S/I; 1138899
dc.information.autorucEscuela de Ingeniería; Gil, Juan Carlos; S/I; 1133691
dc.information.autorucEscuela de Ingeniería; Parra Santander, Denis; 0000-0001-9878-8761; 1011554
dc.language.isoen
dc.nota.accesocontenido parcial
dc.pagina.final134
dc.pagina.inicio123
dc.publisherSpringer, Cham
dc.relation.ispartofRecommender Systems for Sustainability and Social Good: First International Workshop, RecSoGood 2024, Bari, Italy, October 18, 2024, Proceedings
dc.rightsacceso restringido
dc.subjectInformation systems
dc.subjectEcologic recommender systems
dc.subject.ddc000
dc.subject.deweyCiencias de la computaciónes_ES
dc.subject.ods13 Climate action
dc.subject.odspa13 Acción por el clima
dc.title14 Kg of CO2: Analyzing the Carbon Footprint and Performance of Session-Based Recommendation Algorithms
dc.typecomunicación de congreso
sipa.codpersvinculados1138899
sipa.codpersvinculados1133691
sipa.codpersvinculados1011554
sipa.trazabilidadORCID;2025-04-21
Files