{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T17:38:37Z","timestamp":1740159517922,"version":"3.37.3"},"reference-count":21,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,1,17]],"date-time":"2022-01-17T00:00:00Z","timestamp":1642377600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,1,17]],"date-time":"2022-01-17T00:00:00Z","timestamp":1642377600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100009534","name":"Universit\u00e4t Stuttgart","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100009534","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Datenbank Spektrum"],"published-print":{"date-parts":[[2022,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>To obtain accurate predictions of classifiers, model ensembles comprising multiple trained machine learning models are nowadays used. In particular, <jats:italic>dynamic model ensembles<\/jats:italic> pick the most accurate model for each query object, by applying the model that performed best on similar data. Dynamic model ensembles may however suffer, similarly to single machine learning models, from bias, which can eventually lead to unfair treatment of certain groups of a\u00a0general population. To mitigate unfair classification, recent work has thus proposed <jats:italic>fair model ensembles<\/jats:italic>, that instead of focusing (solely) on accuracy also optimize <jats:italic>global fairness<\/jats:italic>. While such global fairness globally minimizes bias, imbalances may persist in different regions of the data, e.g., caused by some local bias maxima leading to <jats:italic>local unfairness<\/jats:italic>.<\/jats:p><jats:p>Therefore, we extend our previous work by including a\u00a0framework that bridges the gap between dynamic model ensembles and fair model ensembles. More precisely, we investigate the problem of devising locally fair and accurate dynamic model ensembles, which ultimately optimize for equal opportunity of similar subjects. We propose a\u00a0general framework to perform this task and present several algorithms implementing the framework components. In this paper we also present a\u00a0runtime-efficient framework adaptation that keeps the quality of the results on a\u00a0similar level. Furthermore, new fairness metrics are presented as well as detailed informations about necessary data preparations.<\/jats:p><jats:p>Our evaluation of the framework implementations and metrics shows that our approach outperforms the state-of-the art for different types and degrees of bias present in training data in terms of both local and global fairness, while reaching comparable accuracy.<\/jats:p>","DOI":"10.1007\/s13222-021-00401-y","type":"journal-article","created":{"date-parts":[[2022,1,17]],"date-time":"2022-01-17T10:03:42Z","timestamp":1642413822000},"page":"23-43","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Metrics and Algorithms for Locally Fair and Accurate Classifications using Ensembles"],"prefix":"10.1007","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6967-1203","authenticated-orcid":false,"given":"Nico","family":"L\u00e4ssig","sequence":"first","affiliation":[]},{"given":"Sarah","family":"Oppold","sequence":"additional","affiliation":[]},{"given":"Melanie","family":"Herschel","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2022,1,17]]},"reference":[{"issue":"9","key":"401_CR1","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1145\/361002.361007","volume":"18","author":"JL Bentley","year":"1975","unstructured":"Bentley JL (1975) Multidimensional binary search trees used for associative searching. Commun ACM 18(9):509\u2013517","journal-title":"Commun ACM"},{"issue":"2","key":"401_CR2","first-page":"4","volume":"8","author":"N Bhatia","year":"2010","unstructured":"Bhatia N (2010) Survey of nearest neighbor techniques. Int J Comput Sci Inf Secur 8(2):4","journal-title":"Int J Comput Sci Inf Secur"},{"issue":"2","key":"401_CR3","doi-asserted-by":"publisher","first-page":"277","DOI":"10.1007\/s10618-010-0190-x","volume":"21","author":"T Calders","year":"2010","unstructured":"Calders T, Verwer S (2010) Three naive bayes approaches for discrimination-free classification. Data Min Knowl Disc 21(2):277\u2013292","journal-title":"Data Min Knowl Disc"},{"unstructured":"Coughlan S (2020) Why did the A\u2011level algorithm say no? https:\/\/www.bbc.com\/news\/education-53787203. Accessed: 16\u00a0December 2021","key":"401_CR4"},{"key":"401_CR5","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1016\/j.inffus.2017.09.010","volume":"41","author":"RM Cruz","year":"2018","unstructured":"Cruz RM, Sabourin R, Cavalcanti GD (2018) Dynamic classifier selection: recent advances and perspectives. Inf Fusion 41:195\u2013216","journal-title":"Inf Fusion"},{"doi-asserted-by":"crossref","unstructured":"Dastin J (2018) Amazon scraps secret AI recruiting tool that showed bias against women. https:\/\/www.reuters.com\/article\/us-amazon-com-jobs-automation-insight\/amazon-scraps-secret-ai-recruiting-tool-thatshowed-bias-against-women-idUSKCN1MK08G. Accessed: 16\u00a0December 2021","key":"401_CR6","DOI":"10.1201\/9781003278290-44"},{"unstructured":"Dua D, Graff C (2019) UCI machine learning repository. http:\/\/archive.ics.uci.edu\/ml. Accessed: 16\u00a0December 2021","key":"401_CR7"},{"key":"401_CR8","first-page":"3723","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV)","author":"N Dvornik","year":"2019","unstructured":"Dvornik N, Schmid C, Mairal J (2019) Diversity with cooperation: ensemble methods for few-shot classification. In: Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), pp 3723\u20133731"},{"key":"401_CR9","first-page":"119","volume-title":"Conference on Fairness, Accountability and Transparency","author":"C Dwork","year":"2018","unstructured":"Dwork C, Immorlica N, Kalai AT et al (2018) Decoupled classifiers for group-fair and efficient machine learning. In: Conference on Fairness, Accountability and Transparency, pp 119\u2013133"},{"key":"401_CR10","first-page":"16","volume":"160907236","author":"SA Friedler","year":"2016","unstructured":"Friedler SA, Scheidegger C, Venkatasubramanian S (2016) On the (im)possibility of fairness. arXiv 160907236:16","journal-title":"arXiv"},{"key":"401_CR11","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1145\/3306618.3317950","volume-title":"Proceedings of the 2019 AAAI\/ACM Conference on AI, Ethics, and Society","author":"S Garg","year":"2019","unstructured":"Garg S, Perot V, Limtiaco N et al (2019) Counterfactual fairness in text classification through robustness. In: Proceedings of the 2019 AAAI\/ACM Conference on AI, Ethics, and Society ACM, AIES \u201919, pp 219\u2013226"},{"key":"401_CR12","volume-title":"Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: concepts, tools, and techniques to build intelligent systems","author":"A G\u00e9ron","year":"2019","unstructured":"G\u00e9ron\u00a0A (2019) Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: concepts, tools, and techniques to build intelligent systems. O\u2019Reilly Media, Sebastopol"},{"key":"401_CR13","first-page":"1","volume-title":"2009 2nd International Conference on Computer, Control and Communication","author":"F Kamiran","year":"2009","unstructured":"Kamiran\u00a0F, Calders\u00a0T (2009) Classifying without discriminating. In: 2009 2nd International Conference on Computer, Control and Communication. IEEE, Karachi, pp\u00a01\u20136"},{"key":"401_CR14","volume-title":"BTW 2021","author":"N L\u00e4ssig","year":"2021","unstructured":"L\u00e4ssig N, Oppold S, Herschel M (2021) Using falces against bias in automated decisions by integrating fairness in dynamic model ensembles. In: BTW 2021"},{"key":"401_CR15","doi-asserted-by":"publisher","first-page":"560","DOI":"10.1145\/1401890.1401959","volume-title":"Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining","author":"D Pedreschi","year":"2008","unstructured":"Pedreschi D, Ruggieri S, Turini F (2008) Discrimination-aware data mining. In: Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining KDD \u201908. ACM Press, Las Vegas, pp\u00a0560\u2013568"},{"issue":"3","key":"401_CR16","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/MCAS.2006.1688199","volume":"6","author":"R Polikar","year":"2006","unstructured":"Polikar R (2006) Ensemble based systems in decision making. IEEE Circuits Syst Mag 6(3):21\u201345","journal-title":"IEEE Circuits Syst Mag"},{"key":"401_CR17","first-page":"4886","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Z Shen","year":"2019","unstructured":"Shen Z, He Z, Xue X (2019) MEAL: Multi-model ensemble via adversarial learning. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol 33, pp 4886\u20134893"},{"issue":"10","key":"401_CR18","doi-asserted-by":"publisher","first-page":"13","DOI":"10.5120\/11430-6785","volume":"67","author":"A Singh","year":"2013","unstructured":"Singh A, Yadav A, Rana A (2013) K\u2011means with three different distance metrics. IJCA 67(10):13\u201317","journal-title":"IJCA"},{"issue":"4","key":"401_CR19","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1109\/34.588027","volume":"19","author":"K Woods","year":"1997","unstructured":"Woods K, Kegelmeyer WP, Bowyer K (1997) Combination of multiple classifiers using local accuracy estimates. IEEE Trans Pattern Anal Machine Intell 19(4):405\u2013410","journal-title":"IEEE Trans Pattern Anal Machine Intell"},{"key":"401_CR20","first-page":"5909","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"H Zheng","year":"2019","unstructured":"Zheng H, Zhang Y, Yang L et al (2019) A\u00a0new ensemble learning framework for 3d biomedical image segmentation. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol 33, pp 5909\u20135916"},{"issue":"4","key":"401_CR21","doi-asserted-by":"publisher","first-page":"1060","DOI":"10.1007\/s10618-017-0506-1","volume":"31","author":"I \u017dliobait\u0117","year":"2017","unstructured":"\u017dliobait\u0117 I (2017) Measuring discrimination in algorithmic decision making. Data Min Knowl Disc 31(4):1060\u20131089","journal-title":"Data Min Knowl Disc"}],"container-title":["Datenbank-Spektrum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13222-021-00401-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13222-021-00401-y\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13222-021-00401-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,3,22]],"date-time":"2022-03-22T08:24:30Z","timestamp":1647937470000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13222-021-00401-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,17]]},"references-count":21,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,3]]}},"alternative-id":["401"],"URL":"https:\/\/doi.org\/10.1007\/s13222-021-00401-y","relation":{},"ISSN":["1618-2162","1610-1995"],"issn-type":[{"type":"print","value":"1618-2162"},{"type":"electronic","value":"1610-1995"}],"subject":[],"published":{"date-parts":[[2022,1,17]]},"assertion":[{"value":"31 October 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 December 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 January 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}