We are pleased to announce that the paper “A Framework towards Computational Narrative Analysis on Blogs” written by Kiran Kumar Bandeli, Muhammad Nihal Hussain, and Nitin Agarwal has been published in The 3rd International Workshop on Narrative Extraction from Texts (Text2Story’20) co-located with 42nd European Conference on Information Retrieval (ECIR 2020).
The paper introduces the idea of extracting narratives automatically on blogs to provide a better situational awareness. Narratives can be represented as semantic triplets of actors and their actions towards other actors.
This is important because it provides sociologists/political scientists to gain situational awareness by tracking different opinions, political views, and narratives as they are shaped. Since blogs are not regulated by any authority and have no restriction on character limit, providing bloggers with not only space for richer content but also serve as a platform for agenda-setting and content framing abetting development of narratives. However, the deluge of posts in the blogosphere makes it impractical to manually identify narratives paving way for the need of computational models.
The research contributions include introducing a novel framework that identifies narratives automatically with the notion of extracting actors/actions using natural language processing techniques and ranking the narratives for a blogpost in the order of their dominance.
The research has far-reaching implications. First, the model developed can help social scientists identify narratives computationally reducing human effort substantially. Second, the results from this research could be used to develop counter narratives to stem propaganda campaigns.
The presentation was delivered virtually and is available here.