Recommender systems are prevalent in many applications, but hide risks; issues like bias propagation have been on the focus of related studies in recent years. My own research revolves around tracking bias in the book recommendation domain. Specifically, I am interested in whether the incorporation of recommender systems in a library’s loaning system serves their social responsibility and purpose, with bias being the main point of concern. To this end, I engage with the topic in three ways; by mapping the area of ethics in book recommendation, by investigating and reflecting on challenges with studying bias in recommender systems in general, and by showcasing a set of social implication of statistical bias in the book recommendation domain in particular. In this doctoral symposium paper, I further elaborate on the problem at hand, the outline of my thesis, the progress I have made so far, as well as my plans for future work along with specific questions that have arisen from my research efforts.

, , ,
doi.org/10.1145/3640457.3688025
18th ACM Conference on Recommender Systems, RecSys 2024
Centrum Wiskunde & Informatica, Amsterdam (CWI), The Netherlands

Daniil, S. (2024). Bias in book recommendation. In Proceedings of the ACM Conference on Recommender Systems (pp. 1376–1381). doi:10.1145/3640457.3688025