Equality of Learning Opportunity in Personalized Recommendations
Preprint
Abstract
Online educational platforms are promising to play a primary role in mediating the success of individuals' careers. Hence, while building overlying content recommendation services, it becomes essential to ensure that learners are provided with equal learning opportunities, according to the platform values, context, and pedagogy. Even though the importance of creating equality of learning opportunities has been well investigated in traditional institutions, how it can be operationalized scalably in online learning ecosystems through recommender systems is still under-explored. In this paper, we formalize principles, that aim to model a range of learning opportunity properties in recommendations, and a metric that combines them to quantify the equality of learning opportunities among learners. Then, we envision a scenario wherein platform owners seek to guarantee that the generated recommendations meet each principle, to a certain degree, for all learners, constrained to their individual preferences. Under this view, we provide observations on learning opportunities in a real-world online dataset, highlighting inequalities among learners. To mitigate this effect, we propose a post-processing approach that balances personalization and learning opportunity equality in recommendations. Experiments on a large-scale dataset demonstrate that our approach leads to higher equality of learning opportunity, with a small loss in personalization.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Designing for serendipity in a university course recommendation system2020
- On the negative impact of social influence in recommender systems: A study of bribery in collaborative hybrid algorithmsInformation Processing & Management, 2020
- Multistakeholder recommendation: Survey and research directionsUser Modeling and User-Adapted Interaction, 2020
- A difficulty ranking approach to personalization in E-learningInternational Journal of Human-Computer Studies, 2019
- FaiRecSys: mitigating algorithmic bias in recommender systemsInternational Journal of Data Science and Analytics, 2019
- What is a virtual world? Definition and classificationEducational Technology Research and Development, 2018
- Continuing Education and Personalization of Design Methods to Improve their Acceptance in Practice – An Explorative StudyProcedia CIRP, 2017
- Updatable, Accurate, Diverse, and Scalable Recommendations for Interactive ApplicationsACM Transactions on Interactive Intelligent Systems, 2016
- Cumulated gain-based evaluation of IR techniquesACM Transactions on Information Systems, 2002
- Constructing Long, Dense Time Series of Inequality Using the Theil Index2001