No NLP Task Should be an Island: Multi-disciplinarity for Diversity in News Recommender Systems

Published in Proceedings of the EACL Hackashop on News Media Content Analysis and Automated Report Generation (co-located at EACL online), 2021

Myrthe Reuver, Antske Fokkens, Suzan Verberne. (2021). “No NLP Task Should be an Island: Multi-disciplinarity for Diversity in News Recommender Systems’ In: Proceedings of the EACL Hackashop on News Media Content Analysis and Automated Report Generation (co-located at EACL 2021, online). Association of Computational Linguistics, p. 45–55.

This position paper with my advisors Antske Fokkens and Suzan Verberne is the result of the first 6 months of my PhD, reading and thinking about my role in the larger project on diversity in news recommendation. I learned about models of democracy, and the role of viewpoint diversity in these models, from my interdisciplinary colleagues. I decided to investigate how different established #NLProc tasks can be connected to the deliberative model of democracy. This work gives me a nice starting point for my experimental work.