A study of top-k measures for discrimination discovery

By Dino Pedreschi, Salvatore Ruggieri, Franco Turini. 2012

58 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

FairSight: Visual Analytics for Fairness in Decision Making (2019, background): “In fair ranking, early studies mainly focused on quantifying the discrimination with proposed ranking measures in the top-k list [32], or indirect discrimination [43].”
A survey on measuring indirect discrimination in machine learning (2015, background): “…discovery typically mines association and classiﬁ-cation rules from the data, and then assesses those rules in terms of potential discrimination [Ruggieri et al. 2010; Romei et al. 2012; Hajian and Domingo-Ferrer 2013; Pedreschi et al. 2012; Luong et al. 2011; Mancuhan and Clifton 2014].”
Certifying and Removing Disparate Impact (2014, background): “[14] D. Pedreschi, S. Ruggieri, and F. Turini.”
Fairness in Machine Learning: A Survey (2020, background): “…fall into three distinct areas: (1) fair clustering (e.g., References [7, 8, 15, 22, 23, 62, 67, 179, 248, 258, 269]); (2) investigating the presence and detection of discrimination in association rule mining (e.g., References [136, 230, 231]); and (3) transfer learning (e.g., References [78, 90]).”
Fairness in Relational Domains (2018, methodology): “Existing data-driven fairness measures [19] can be defined in terms of p 1 and p 2 based on our relational notions as follows: (1) Risk difference : RD = p 1 − p 2 , also known as absolute risk reduction.”
Optimized Pre-Processing for Discrimination Prevention (2017, background): “” The combination of (3) and (1) generalizes the extended lift criterion proposed in the literature [Pedreschi et al., 2012], while the combination of (3) and (2) generalizes selective and contrastive lift.”
Discrimination-aware data mining: a survey (2017, methodology): “The terminologies as defined by Ruggieri et al. (2010a) and Pedreschi et al.”
Optimized Data Pre-Processing for Discrimination Prevention (2017, methodology): “For this work, we choose J to be the following probability ratio measure: The combination of (3) and (1) generalizes the extended lift criterion proposed in the literature (Pedreschi et al., 2012), while the combination of (3) and (2) generalizes selective and contrastive lift.”
Obtaining Fairness using Optimal Transport Theory (2018, background): “…over the Internet, they are now widely used in a large number of very sensitive areas such as medicine, human ressources with hiring policies, banks and insurance (lending), police, and justice with criminal sentencing, see for instance in [BHJ + 17] [PRT12] or [FSV + 18] and references therein.”
Better decision support through exploratory discrimination-aware data mining: foundations and empirical evidence (2014, methodology): “The factors setting (Bank, ADA as short for anti-discrimination agency) and mining form (eDADM, cDADM, DM) were manipulated between subjects.”
Addressing multiple metrics of group fairness in data-driven decision making (2020, background): “A protected group is a group distinguished by a protected feature [26], where protected features are usually categories from a given legal context, such as gender or race in anti-discrimination legislation.”
Exploring Discrimination: A User-centric Evaluation of Discrimination-Aware Data Mining (2012, background): “Note that agreed-upon deﬁnitions of “bad patterns” are still being developed, cf. [8], [11].”
A multidisciplinary survey on discrimination analysis (2013, methodology): “D risk plane can be dramatically different (Pedreschi et al., 2012).”
Measuring discrimination in algorithmic decision making (2017, methodology): “A data mining approach for discrimination discovery typically extracts association and classification rules from data, and then evaluates those rules in terms of potential discrimination (Hajian and Domingo-Ferrer 2013; Luong et al. 2011; Mancuhan and Clifton 2014; Pedreschi et al. 2012; Romei et al”