e-Fold Cross-Validation for Recommender-System Evaluation

e-Fold Cross-Validation for Recommender-System Evaluation. Baumgart, M., Wegmeth, L., Vente, T., & Beel, J. In First International Workshop on Recommender Systems for Sustainability and Social Good (RecSoGood), October, 2024.

Paper abstract bibtex

To combat the rising energy consumption of recommender systems we implement a novel alternative for k-fold cross validation. This alternative, named e-fold cross validation, aims to minimize the number of folds to achieve a reduction in power usage while keeping the reliability and robustness of the test results high. We tested our method on 5 recommender system algorithms across 6 datasets and compared it with 10-fold cross validation. On average e-fold cross validation only needed 41.5% of the energy that 10-fold cross validation would need, while it’s results only differed by 1.81%. We conclude that e-fold cross validation is a promising approach that has the potential to be an energy efficient but still reliable alternative to k-fold cross validation.

@inproceedings{baumgart_e-fold_2024,
	title = {e-{Fold} {Cross}-{Validation} for {Recommender}-{System} {Evaluation}},
	url = {https://isg.beel.org/pubs/2024-e-folds-recsys-baumgart.pdf},
	abstract = {To combat the rising energy consumption of recommender systems we implement a novel alternative for k-fold cross validation. This alternative, named e-fold cross validation, aims to minimize the number of folds to achieve a reduction in power usage while keeping the reliability and robustness of the test results high. We tested our method on 5 recommender system algorithms across 6 datasets and compared it with 10-fold cross validation. On average e-fold cross validation only needed 41.5\% of the energy that 10-fold cross validation would need, while it’s results only differed by 1.81\%. We conclude that e-fold cross validation is a promising approach that has the potential to be an energy efficient but still reliable alternative to k-fold cross validation.},
	language = {en},
	booktitle = {First {International} {Workshop} on {Recommender} {Systems} for {Sustainability} and {Social} {Good} ({RecSoGood})},
	author = {Baumgart, Moritz and Wegmeth, Lukas and Vente, Tobias and Beel, Joeran},
	month = oct,
	year = {2024},
	keywords = {⛔ No DOI found},
}

Downloads: 0

{"_id":"eRftYo8udbqCHXmJr","bibbaseid":"baumgart-wegmeth-vente-beel-efoldcrossvalidationforrecommendersystemevaluation-2024","author_short":["Baumgart, M.","Wegmeth, L.","Vente, T.","Beel, J."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"e-Fold Cross-Validation for Recommender-System Evaluation","url":"https://isg.beel.org/pubs/2024-e-folds-recsys-baumgart.pdf","abstract":"To combat the rising energy consumption of recommender systems we implement a novel alternative for k-fold cross validation. This alternative, named e-fold cross validation, aims to minimize the number of folds to achieve a reduction in power usage while keeping the reliability and robustness of the test results high. We tested our method on 5 recommender system algorithms across 6 datasets and compared it with 10-fold cross validation. On average e-fold cross validation only needed 41.5% of the energy that 10-fold cross validation would need, while it’s results only differed by 1.81%. We conclude that e-fold cross validation is a promising approach that has the potential to be an energy efficient but still reliable alternative to k-fold cross validation.","language":"en","booktitle":"First International Workshop on Recommender Systems for Sustainability and Social Good (RecSoGood)","author":[{"propositions":[],"lastnames":["Baumgart"],"firstnames":["Moritz"],"suffixes":[]},{"propositions":[],"lastnames":["Wegmeth"],"firstnames":["Lukas"],"suffixes":[]},{"propositions":[],"lastnames":["Vente"],"firstnames":["Tobias"],"suffixes":[]},{"propositions":[],"lastnames":["Beel"],"firstnames":["Joeran"],"suffixes":[]}],"month":"October","year":"2024","keywords":"⛔ No DOI found","bibtex":"@inproceedings{baumgart_e-fold_2024,\n\ttitle = {e-{Fold} {Cross}-{Validation} for {Recommender}-{System} {Evaluation}},\n\turl = {https://isg.beel.org/pubs/2024-e-folds-recsys-baumgart.pdf},\n\tabstract = {To combat the rising energy consumption of recommender systems we implement a novel alternative for k-fold cross validation. This alternative, named e-fold cross validation, aims to minimize the number of folds to achieve a reduction in power usage while keeping the reliability and robustness of the test results high. We tested our method on 5 recommender system algorithms across 6 datasets and compared it with 10-fold cross validation. On average e-fold cross validation only needed 41.5\\% of the energy that 10-fold cross validation would need, while it’s results only differed by 1.81\\%. We conclude that e-fold cross validation is a promising approach that has the potential to be an energy efficient but still reliable alternative to k-fold cross validation.},\n\tlanguage = {en},\n\tbooktitle = {First {International} {Workshop} on {Recommender} {Systems} for {Sustainability} and {Social} {Good} ({RecSoGood})},\n\tauthor = {Baumgart, Moritz and Wegmeth, Lukas and Vente, Tobias and Beel, Joeran},\n\tmonth = oct,\n\tyear = {2024},\n\tkeywords = {⛔ No DOI found},\n}\n\n","author_short":["Baumgart, M.","Wegmeth, L.","Vente, T.","Beel, J."],"key":"baumgart_e-fold_2024","id":"baumgart_e-fold_2024","bibbaseid":"baumgart-wegmeth-vente-beel-efoldcrossvalidationforrecommendersystemevaluation-2024","role":"author","urls":{"Paper":"https://isg.beel.org/pubs/2024-e-folds-recsys-baumgart.pdf"},"keyword":["⛔ No DOI found"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/6655/collections/3TB3KT36/items?key=VFvZhZXIoHNBbzoLZ1IM2zgf&format=bibtex&limit=100","dataSources":["7KNAjxiv2tsagmbgY"],"keywords":["⛔ no doi found"],"search_terms":["fold","cross","validation","recommender","system","evaluation","baumgart","wegmeth","vente","beel"],"title":"e-Fold Cross-Validation for Recommender-System Evaluation","year":2024}