Media framing refers to the deliberate presentation of information in order to elicit a desired response or shift in reader’s attitude. While there is an expanding body of research in Natural Language Processing (NLP) aimed at automatically identifying framing in news or social media, a majority of these efforts rely on straightforward yet somewhat reductionist methods, such as topic modeling or basic classifiers. These approaches tend to formalize the task as single-label classification, often neglecting the co-existence and interplay of various framing perspectives. This project focuses on designing novel unsupervised machine learning techniques for identifying media frames from large collections of text (news , social media, public relation communication etc) in a scalable manner.
Mohammad Ali, Naeemul Hassan. Semantic-based unsupervised framing analysis (sufa): a novel approach for computational framing analysis. Association for Education in Journalism and Mass Communication (AEJMC). 2023
Mohammad Ali, Naeemul Hassan. A survey of computational framing analysis approaches. In Proceedings of the 2022 Empirical Methods in Natural Language Processing. 2022