Bitan Majumder

Bitan Majumder

About Me

I am a Doctoral student of Computer Science at VSAC, Ashoka University, working with Dr. Anirban Sen. Prior to joining the Department of Computer science at Ashoka, I was a Graduate student at Pondicherry Central University. I did my undergraduate studies at Ramakrishna Mission Vivekananda Centenary College. My research interest is in the field of Computational Social Science and Natural Language Processing. I did my schooling at Ramakrishna Mission Vidyapith, Purulia. My hobby is reading story books(short stories, historical fiction) and watching football matches(culé).

Research Interest

Computational Social Science, Natural Language Processing, Social Network Analysis

Experience

Jan, 2026- April, 2026 : Research Assistant at Ashoka University, Sonipat.
Jan, 2025 – Dec, 2025 : Intern at Ashoka University, Sonipat.
June, 2024 – Sept, 2024 : Intern at IEEE Computer Society Student Branch Chapter of IIT Kharagpur. Worked on Time Series Modeling and Context-aware Generative AI.
June, 2023 – Dec, 2023 : Intern at Department of Computer Science and Technology, IIEST Shibpur, under the guidance of Dr. Asit Kumar Das. Worked on Natural Language Processing based answer script vectorization.
Sept, 2022 – Feb, 2023 : Intern at Department of Computer Science, West Bengal State University, under the guidance of Dr. Kaushik Roy. I was part of a team that developed an Automated Bangla character extraction technique from isolated data collection forms.

Project

Code-mixed Sarcasm Detection of Large Language Models

Large Language Models are capable of doing almost everything that we need in our day-to-day lives, yet they perform poorly when it comes to basic tasks such as sarcasm detection in code-mixed sentences. Comparatively a transformer-based language model such as DistilBERT performs quite well if it is fine-tuned on code-mixed sarcasm data in detection tasks.

Automated Answer Script Evaluation

Answer script evaluation is a very tedious job which is subject to various human biases. This project was aimed to develop an automated system based on NLP to predict score for answers to broad-answer-type questions.