Social & Computational Intelligence Research

Computational Linguistics, Machine Learning, Sociology, and Communication unite to build the next generation of Human-Centered AI

$1.079M in research funding
4 doctoral researchers
43 M.S. students
2 open dataset resources
3 collaborating universities

Student involvement in research is a core value of our group at SCIRE. Of the M.S. and undergraduate students who have worked in our group, 19 have co-authored peer-reviewed publications at venues including ACL, EMNLP, IEEE ICHI, and others.

Ritwik Banerjee

Dr. Ritwik Banerjee

Principal Investigator, SCIRE Group

206 New Computer Science
100 Engineering Drive
Stony Brook, New York

SCIRE (Latin: “to know”) explores and advances the frontiers of computational linguistics and its applications in contemporary society, designing and developing machine learning algorithms and models that promise unprecedented comprehension of natural language pragmatics. We investigate social problems expressed using human language, especially problems that rely on a careful analysis of context beyond the immediate linguistic neighborhood of such expressions, including the use of figurative language, e.g., fallacious argumentation by means of deflection, the use of subtle bias, and the use of perceptions created through rhetoric and dialectic maneuvers. Pushing these boundaries may enable rich applications across multiple domains such as computational journalism, representation learning, communication theory, healthcare informatics, human-computer interaction, and information retrieval. To develop these cutting-edge AI technology vectors, our research lies at the intersection of computational linguistics, machine learning, sociology, and psychology.

Our core research areas are natural language pragmatics, semantic representation, information retrieval, and computational argumentation, which are explored through foundational theories in sociology, psychology, mathematics, and linguistics.

 

News & Announcements

05.2026. publication Our multilingual (English, Russian, and Mandarin Chinese), multinational (USA, UK, Ukraine, Russia, and China), and longitudinal (31 months) corpus on the Russo-Ukrainian war is published in LREC 2026.  
The corpus is dubbed dnipro — Diverse Narratives and International Perspectives on the Russo-Ukrainian Offensive — and released as a public resource on Zenodo:
  •   https://doi.org/10.5281/zenodo.20433469
01.2026. phd Chenlu Wang defends her Ph.D. on Pragmatic Language Understanding and Information Integrity. Dr. Wang will begin her next chapter as a Research Scientist at Meta..
12.2025. publication Two publications at IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2025:
  On evidence-based reasoning across languages  
  On medical expert identification  
Collaborators: Chaoyuan Zuo, Nankai University
09.2025. award We are awarded a research grant by the Society of Family Planning to study misinformation and disinformation surrounding contraception.
09.2025. publication Vishnu Raja’s work on dysarthric speech @ EMNLP 2025.   
Collaborators: H. Andrew Schwartz @ Vanderbilt University