The Artificial Intelligence and Linguistics Program is an initiative that operates at the intersection of discourse analysis, socio-pragmatics, and media linguistics. Our primary objective is to investigate the various aspects of digital discourse, including its contextual construction, sustainability, and potential perlocutionary consequences, particularly from socio-psychological, real-world harm, and national security perspectives. Given the immense velocity, volume, and linguistic diversity that exists in data generated across multiple languages on social media, manual intervention and real-time analysis for digital discourse are impractical. We are committed to the development of artificial intelligence (AI) systems for natural language processing (NLP) and natural language understanding (NLU) within this program. Our ultimate goal is to leverage these AI-driven insights and transform them into actionable solutions that can help formulate policies to effectively address the complexities of digital discourse.
The SpeeD-IL Project: Speech Datasets and Models for Indian Languages
The SpeeD-IL project, in collaboration with UnReaL-TecE LLP and Karya Inc., aims
POLING – Political Opinion Analysis using Linguistic Data
The POLING project applies computational linguistics to analyze social media for political sentiment in
Communal and Misogynistic Aggression in Hindi-English-Bangla (The ComMA Project)
This project, in partnership with Dr. Bhimrao Ambedkar University and UnReaL-TecE LLP, studies
Measuring Harm Potential of Online Hate Speech
This project is developing a framework to assess the ‘harm potential’ of online posts in