Research Dimensions

5 Projects

Cyber informatics

We design models for detecting and mitigating various cybercrimes on online social media, including the spread of fake news, misinformation, fraud activities, black-market driven collusion, hate and offensive speech, and cyberbullying. Our approach leverages advanced machine learning techniques to enhance the accuracy and effectiveness of these models.

4 Projects

Conversational Dialogues

We regard various modules of a dialog system, such as intent classification, slot-filling, dialog-act classification, dialog management, and dialog generation, for both goal-oriented and generic conversational chatbots. Moreover, we inspect conversational dialogs for various supplementary tasks as well, e.g., speaker profiling, emotion-flip reasoning, etc.

4 Projects

Graph | Complex Networks

We study social, information, and scientific networks to explore various structural and behavioral properties of nodes and edges. Our research aims to understand the dynamics of these networks and how they evolve over time. We utilize advanced computational methods and data analytics to uncover hidden patterns within complex systems.

3 Projects

Multimodality

We study the interdisciplinary domain of speech and/or vision modalities with the textual processing for a number of traditional and recent natural language processing tasks such as summarization, opinion mining, emotion analysis, sarcasm detection, offensive post detection, etc. This holistic approach provides deeper insights and more effective solutions across a variety of applications.

3 Projects

Code-mixed and Low Resource Language Processing

We explore a wide range of natural language processing tasks in the vicinity of code-mixed and low-resource languages in the Indian context. The prime objective is to facilitate the technology to a vast non-native English speakers and also make the applications widely available.

3 Projects

AI for MHx (Manashi)

We are focused on building human-centered artificial intelligence tools to support mental healthcare. We aim to combine clinical insight, multilingual natural language processing (NLP), and ethical real-world deployment to create AI systems that help clinicians with tasks like screening, history taking, assessment, documentation, and other aspects of mental health care delivery.