Discrimination Data Analysis: A Multi-disciplinary Bibliography

Romei, Andrea; Ruggieri, Salvatore

doi:10.1007/978-3-642-30487-3_6

Cited by 10 publications

(5 citation statements)

References 233 publications

(176 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…unjustly on grounds of race or colour or sex" [2]; a multi-disciplinary overview is provided in [3]. Such "unjust" grounds are legally codified in many countries and may include further characteristics.…”

Section: A Data Mining and Discriminationmentioning

confidence: 99%

Exploring Discrimination: A User-centric Evaluation of Discrimination-Aware Data Mining

Berendt

Preibusch

2012

2012 IEEE 12th International Conference on Data Mining Workshops

View full text Add to dashboard Cite

Abstract-Discrimination-aware data mining (DADM) aims at deriving patterns that do not discriminate on "unjust grounds" such as gender, ethnicity or nationality. DADM safeguards can be very helpful for decision-support applications in fields such as banking or employment. However, constraining data mining to exclude a fixed enumeration of potentially discriminatory features is too restrictive. It should be complemented by exploratory DADM. We discuss these two forms of DADM and their requirements for evaluation, and we discuss and refine our DCUBE-GUI tool as a system for exploratory DADM. In a user study administered via Mechanical Turk, we show that tools such as DCUBE-GUI can successfully assist novice users in exploring discrimination in data mining.

show abstract

Section: A Data Mining and Discriminationmentioning

confidence: 99%

Exploring Discrimination: A User-centric Evaluation of Discrimination-Aware Data Mining

Berendt

Preibusch

2012

2012 IEEE 12th International Conference on Data Mining Workshops

View full text Add to dashboard Cite

show abstract

“…Empirical research in the field of labor market discrimination usually applies statistical inference techniques to data sets with two possible purposes: (i) Testing the consequences of discrimination described by various theoretical economic models, or (ii) evaluating the contribution of different types of discrimination to different treatments on the labor market [16]. The most commonly used methods of discrimination analysis are statistical tests for means and proportions, linear (generalized or logistic) regression models and other econometric models.…”

Section: Literature Reviewmentioning

confidence: 99%

Investigating Labor Market Discrimination in Romania

2020

View full text Add to dashboard Cite

An inclusive labor market, with equal opportunities for all, is a desideratum of all states, and one of the Sustainable Development Goals established by the United Nations for the year 2030. Discrimination can take many forms, based on many reasons, and has a strong impact on the living standards of individuals. Discrimination in the labor market can negatively affect economic performances. The aim of the research is to investigate the perception of individuals regarding discrimination, with a focus on the characteristics of the individuals who feel discriminated in Romania in a work-related context. In order to capture the relevant aspects, we turned to descriptive analysis and logistic regression. The analysis revealed that around 14% of the respondents participating in the survey faced a discriminating situation in a work-related context during the last year. Moreover, less than half of these individuals have taken any action against the discrimination act, the others believing that nothing would have changed. The logistic regression results indicated that Roma people have 6.7 times higher chances to be discriminated in a work-related context as compared to a Romanian individual. Moreover, higher educated individuals are more likely to experience discrimination in the labor market.

show abstract

“…Automated scoring systems may offer some advantages over humans, e.g., higher score consistency (Williamson et al, 2012). Yet like any other machine learning algorithm, models used for score prediction may inadvertently encode discrimination into their decisions due to biases or other imperfections in the training data, spurious correlations, and other factors (Xi, 2010;Romei and Ruggieri, 2013;von Davier, 2016;Zieky, 2016). Given that many such systems are used to score high-stakes standardized tests, the consequences of any form of bias can have a significant effect on people's lives.…”

Section: Motivationmentioning

confidence: 99%

User-centered & Robust NLP OSS: Lessons Learned from Developing & Maintaining RSMTool

Madnani¹,

Loukina²

2020

Proceedings of Second Workshop for NLP Open Source Software (NLP-OSS)

View full text Add to dashboard Cite

For the last 5 years, we have developed and maintained RSMTool -an open-source tool for evaluating NLP systems that automatically score written and spoken responses. RSMTool is designed to be cross-disciplinary, borrowing heavily from NLP, machine learning, and educational measurement. Its crossdisciplinary nature has required us to learn a user-centered development approach in terms of both design and implementation. We share some of these lessons in this paper.

show abstract

Discrimination Data Analysis: A Multi-disciplinary Bibliography

Cited by 10 publications

References 233 publications

Exploring Discrimination: A User-centric Evaluation of Discrimination-Aware Data Mining

Exploring Discrimination: A User-centric Evaluation of Discrimination-Aware Data Mining

Investigating Labor Market Discrimination in Romania

User-centered & Robust NLP OSS: Lessons Learned from Developing & Maintaining RSMTool

Contact Info

Product

Resources

About