Springe direkt zu Inhalt

Bachelor Theses: "Evaluating fairness measures in student performance prediction"

Description

Predicting students’ academic performance, e.g., grades, the success of courses is one of the main topics of educational data mining [1]. Predictive models have a good forecasting quality but the prediction can be biased w.r.t. some protected attributes, such as gender, race, etc. Fairness in machine learning receives increasing attention [2] and despite the many definitions of fairness still there is no consensus on which measure is suitable for which application. In the educational domain, several studies  [3, 4, 5, 6, 7] evaluate fairness in student performance prediction tasks using measures such as: ABROCA, Equalized Odds, etc. But, there is a lack of an overall evaluation of fairness measures for the education domain.

The goal of this bachelor thesis is to investigate the different fairness measures for student performance prediction problems. A comparative analysis of the different measures including experimental evaluation is required.

 An ideal candidate should be:

  • a self-motivated and independent learner
  • knowledgeable about machine learning (indicated by good grades in related courses)
  • experienced with Python The thesis will be supervised by Prof. Eirini Ntoutsi (eirini.ntoutsi@fu-berlin.de) and PhD candidate Tai Le Quy (tai.lequy@fu-berlin.de) from the Institute of Computer Science.References

[1] Shahiri, Amirah Mohamed, and Wahidah Husain. "A review on predicting student's performance using data mining techniques." Procedia Computer Science 72 (2015): 414-422.

[2] Ntoutsi, Eirini, et al. "Bias in data‐driven artificial intelligence systems—An introductory survey." Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 10.3 (2020): e1356.

[3] Bøyum, Steinar. "Fairness in education–a normative analysis of OECD policy documents." Journal of Education Policy 29.6 (2014): 856-870.

[4] Gardner, Josh, Christopher Brooks, and Ryan Baker. "Evaluating the fairness of predictive student models through slicing analysis." Proceedings of the 9th International Conference on Learning Analytics & Knowledge. 2019.

[5] Riazy, Shirin, Katharina Simbeck, and Vanessa Schreck. "Fairness in Learning Analytics: Student At-risk Prediction in Virtual Learning Environments." CSEDU (1). 2020.

[6] Kizilcec, René F., and Hansol Lee. "Algorithmic Fairness in Education." arXiv preprint arXiv:2007.05443 (2020).

[7] Lee, Hansol, and René F. Kizilcec. "Evaluation of Fairness Trade-offs in Predicting Student Success." arXiv preprint arXiv:2007.00088 (2020).