Resume
Download my latest resume to learn more about my experience, skills, and education.
Download Resume (PDF)Research Papers
My academic research and publications in machine learning, natural language processing, and related fields.
MuRFE: A Dataset and Benchmark for Multi-valued Relational Facts Extraction
Pratik Saini, Bhavya Jain, Suchandra Chakraborty, Chayan Sarkar, Tapas Nayak
Under Review at CoNLL 2026
Created synthetic datasets for multi-valued relation extraction using advanced prompt engineering with Llama-3.3-70B and trained sequence tagging with pointer-network models for complex relational fact extraction.
Self-State Evidence Extraction and Well-Being Prediction from Social Media Timelines
Suchandra Chakraborty, Sudeshna Jana, Manjira Sinha, Tirthankar Dasgupta
CLPsych 2025 (ACL Workshop)
Developed RoBERTa-based evidence extraction and retrieval-augmented LLM models for mental health NLP from Reddit posts, achieving top-tier results in identifying adaptive and maladaptive self-states and predicting well-being scores.
Detection and Extraction of Food Recalls and Foodborne Disease Outbreaks using Linguistically-Informed Language Models
Suchandra Chakraborty, Diya Saha, Sudeshna Jana
SMM4H 2025 (AAAI Workshop)
Employed transformer-based architectures and LLMs using few-shot and zero-shot prompt engineering for sentence-level classification and entity extraction from FDA press releases, achieving 92.3% F1 in classification and second-highest accuracy in both subtasks.
What Drives the Variation of Developer Communication Characteristics over Time? An Empirical Study Across Multiple Datasets
Suchandra Chakraborty, Ankan Basu, Aritra Saha, Ishita Bardhan, Subhajit Datta, Subhashis Majumder
COMSYS 2024, Springer LNNS
Applied Temporal Exponential Random Graph Models (TERGM, ERGM, BTERGM) to analyze developer interaction patterns across OpenStack and Eclipse datasets, identifying PageRank as a pivotal factor in network dynamics.
Generation of Data for Training Retinal Image Segmentation Models
Srinjoy Bhuiya, Suchandra Chakraborty, Subhopriyo Sadhukhan, Deba Prasad Mandal, Dinabandhu Bhandari
PReMI 2023, Springer LNCS
Developed novel data augmentation using StyleGAN2 and Pix2Pix to create synthetic retinal image datasets, significantly improving supervised biomedical image segmentation performance over traditional techniques.