Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness & Satisfaction in Recommendation Systems. Mehrotra, R., McInerney, J., Bouchard, H., Lalmas, M., & Diaz, F. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management, of CIKM '18, pages 2243–2251. ACM. event-place: Torino, Italy![Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness & Satisfaction in Recommendation Systems [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Two-sided marketplaces are platforms that have customers not only on the demand side (e.g. users), but also on the supply side (e.g. retailer, artists). While traditional recommender systems focused specifically towards increasing consumer satisfaction by providing relevant content to consumers, two-sided marketplaces face the problem of additionally optimizing for supplier preferences, and visibility. Indeed, the suppliers would want afair opportunity to be presented to users. Blindly optimizing for consumer relevance may have a detrimental impact on supplier fairness. Motivated by this problem, we focus on the trade-off between objectives of consumers and suppliers in the case of music streaming services, and consider the trade-off betweenrelevance of recommendations to the consumer (i.e. user) andfairness of representation of suppliers (i.e. artists) and measure their impact on consumersatisfaction. We propose a conceptual and computational framework using counterfactual estimation techniques to understand, and evaluate different recommendation policies, specifically around the trade-off between relevance and fairness, without the need for running many costly A/B tests. We propose a number of recommendation policies which jointly optimize relevance and fairness, thereby achieving substantial improvement in supplier fairness without noticeable decline in user satisfaction. Additionally, we consider user disposition towards fair content, and propose a personalized recommendation policy which takes into account consumer's tolerance towards fair content. Our findings could guide the design of algorithms powering two-sided marketplaces, as well as guide future research on sophisticated algorithms for joint optimization of user relevance, satisfaction and fairness.
@inproceedings{mehrotra_towards_2018,
location = {New York, {NY}, {USA}},
title = {Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness \& Satisfaction in Recommendation Systems},
isbn = {978-1-4503-6014-2},
url = {http://doi.acm.org/10.1145/3269206.3272027},
doi = {10.1145/3269206.3272027},
series = {{CIKM} '18},
shorttitle = {Towards a Fair Marketplace},
abstract = {Two-sided marketplaces are platforms that have customers not only on the demand side (e.g. users), but also on the supply side (e.g. retailer, artists). While traditional recommender systems focused specifically towards increasing consumer satisfaction by providing relevant content to consumers, two-sided marketplaces face the problem of additionally optimizing for supplier preferences, and visibility. Indeed, the suppliers would want afair opportunity to be presented to users. Blindly optimizing for consumer relevance may have a detrimental impact on supplier fairness. Motivated by this problem, we focus on the trade-off between objectives of consumers and suppliers in the case of music streaming services, and consider the trade-off betweenrelevance of recommendations to the consumer (i.e. user) andfairness of representation of suppliers (i.e. artists) and measure their impact on consumersatisfaction. We propose a conceptual and computational framework using counterfactual estimation techniques to understand, and evaluate different recommendation policies, specifically around the trade-off between relevance and fairness, without the need for running many costly A/B tests. We propose a number of recommendation policies which jointly optimize relevance and fairness, thereby achieving substantial improvement in supplier fairness without noticeable decline in user satisfaction. Additionally, we consider user disposition towards fair content, and propose a personalized recommendation policy which takes into account consumer's tolerance towards fair content. Our findings could guide the design of algorithms powering two-sided marketplaces, as well as guide future research on sophisticated algorithms for joint optimization of user relevance, satisfaction and fairness.},
pages = {2243--2251},
booktitle = {Proceedings of the 27th {ACM} International Conference on Information and Knowledge Management},
publisher = {{ACM}},
author = {Mehrotra, Rishabh and {McInerney}, James and Bouchard, Hugues and Lalmas, Mounia and Diaz, Fernando},
urldate = {2019-07-17},
date = {2018},
note = {event-place: Torino, Italy},
keywords = {fairness, marketplace, satisfaction}
}
Downloads: 0
{"_id":"Ssy9PXMQprZpqAZgu","bibbaseid":"mehrotra-mcinerney-bouchard-lalmas-diaz-towardsafairmarketplacecounterfactualevaluationofthetradeoffbetweenrelevancefairnesssatisfactioninrecommendationsystems","authorIDs":[],"author_short":["Mehrotra, R.","McInerney, J.","Bouchard, H.","Lalmas, M.","Diaz, F."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","location":"New York, NY, USA","title":"Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness & Satisfaction in Recommendation Systems","isbn":"978-1-4503-6014-2","url":"http://doi.acm.org/10.1145/3269206.3272027","doi":"10.1145/3269206.3272027","series":"CIKM '18","shorttitle":"Towards a Fair Marketplace","abstract":"Two-sided marketplaces are platforms that have customers not only on the demand side (e.g. users), but also on the supply side (e.g. retailer, artists). While traditional recommender systems focused specifically towards increasing consumer satisfaction by providing relevant content to consumers, two-sided marketplaces face the problem of additionally optimizing for supplier preferences, and visibility. Indeed, the suppliers would want afair opportunity to be presented to users. Blindly optimizing for consumer relevance may have a detrimental impact on supplier fairness. Motivated by this problem, we focus on the trade-off between objectives of consumers and suppliers in the case of music streaming services, and consider the trade-off betweenrelevance of recommendations to the consumer (i.e. user) andfairness of representation of suppliers (i.e. artists) and measure their impact on consumersatisfaction. We propose a conceptual and computational framework using counterfactual estimation techniques to understand, and evaluate different recommendation policies, specifically around the trade-off between relevance and fairness, without the need for running many costly A/B tests. We propose a number of recommendation policies which jointly optimize relevance and fairness, thereby achieving substantial improvement in supplier fairness without noticeable decline in user satisfaction. Additionally, we consider user disposition towards fair content, and propose a personalized recommendation policy which takes into account consumer's tolerance towards fair content. Our findings could guide the design of algorithms powering two-sided marketplaces, as well as guide future research on sophisticated algorithms for joint optimization of user relevance, satisfaction and fairness.","pages":"2243–2251","booktitle":"Proceedings of the 27th ACM International Conference on Information and Knowledge Management","publisher":"ACM","author":[{"propositions":[],"lastnames":["Mehrotra"],"firstnames":["Rishabh"],"suffixes":[]},{"propositions":[],"lastnames":["McInerney"],"firstnames":["James"],"suffixes":[]},{"propositions":[],"lastnames":["Bouchard"],"firstnames":["Hugues"],"suffixes":[]},{"propositions":[],"lastnames":["Lalmas"],"firstnames":["Mounia"],"suffixes":[]},{"propositions":[],"lastnames":["Diaz"],"firstnames":["Fernando"],"suffixes":[]}],"urldate":"2019-07-17","date":"2018","note":"event-place: Torino, Italy","keywords":"fairness, marketplace, satisfaction","bibtex":"@inproceedings{mehrotra_towards_2018,\n\tlocation = {New York, {NY}, {USA}},\n\ttitle = {Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness \\& Satisfaction in Recommendation Systems},\n\tisbn = {978-1-4503-6014-2},\n\turl = {http://doi.acm.org/10.1145/3269206.3272027},\n\tdoi = {10.1145/3269206.3272027},\n\tseries = {{CIKM} '18},\n\tshorttitle = {Towards a Fair Marketplace},\n\tabstract = {Two-sided marketplaces are platforms that have customers not only on the demand side (e.g. users), but also on the supply side (e.g. retailer, artists). While traditional recommender systems focused specifically towards increasing consumer satisfaction by providing relevant content to consumers, two-sided marketplaces face the problem of additionally optimizing for supplier preferences, and visibility. Indeed, the suppliers would want afair opportunity to be presented to users. Blindly optimizing for consumer relevance may have a detrimental impact on supplier fairness. Motivated by this problem, we focus on the trade-off between objectives of consumers and suppliers in the case of music streaming services, and consider the trade-off betweenrelevance of recommendations to the consumer (i.e. user) andfairness of representation of suppliers (i.e. artists) and measure their impact on consumersatisfaction. We propose a conceptual and computational framework using counterfactual estimation techniques to understand, and evaluate different recommendation policies, specifically around the trade-off between relevance and fairness, without the need for running many costly A/B tests. We propose a number of recommendation policies which jointly optimize relevance and fairness, thereby achieving substantial improvement in supplier fairness without noticeable decline in user satisfaction. Additionally, we consider user disposition towards fair content, and propose a personalized recommendation policy which takes into account consumer's tolerance towards fair content. Our findings could guide the design of algorithms powering two-sided marketplaces, as well as guide future research on sophisticated algorithms for joint optimization of user relevance, satisfaction and fairness.},\n\tpages = {2243--2251},\n\tbooktitle = {Proceedings of the 27th {ACM} International Conference on Information and Knowledge Management},\n\tpublisher = {{ACM}},\n\tauthor = {Mehrotra, Rishabh and {McInerney}, James and Bouchard, Hugues and Lalmas, Mounia and Diaz, Fernando},\n\turldate = {2019-07-17},\n\tdate = {2018},\n\tnote = {event-place: Torino, Italy},\n\tkeywords = {fairness, marketplace, satisfaction}\n}\n\n","author_short":["Mehrotra, R.","McInerney, J.","Bouchard, H.","Lalmas, M.","Diaz, F."],"key":"mehrotra_towards_2018","id":"mehrotra_towards_2018","bibbaseid":"mehrotra-mcinerney-bouchard-lalmas-diaz-towardsafairmarketplacecounterfactualevaluationofthetradeoffbetweenrelevancefairnesssatisfactioninrecommendationsystems","role":"author","urls":{"Paper":"http://doi.acm.org/10.1145/3269206.3272027"},"keyword":["fairness","marketplace","satisfaction"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://fair-ia.ekstrandom.net/fair-ia.bib","creationDate":"2020-04-09T18:53:45.430Z","downloads":0,"keywords":["fairness","marketplace","satisfaction"],"search_terms":["towards","fair","marketplace","counterfactual","evaluation","trade","between","relevance","fairness","satisfaction","recommendation","systems","mehrotra","mcinerney","bouchard","lalmas","diaz"],"title":"Towards a Fair Marketplace: Counterfactual Evaluation of the Trade-off Between Relevance, Fairness & Satisfaction in Recommendation Systems","year":null,"dataSources":["FRCCaPECNMucjb6Hk"]}