Mr. Harijit Tripura

Assistant Professor

Mr. Harijit Tripura joined the Department of Computer Science and Engineering at Dayananda Sagar University, Bengaluru, as Assistant Professor. His work centers on applying machine learning and deep learning techniques to two very different problem spaces: medical image diagnostics and natural language processing for low-resource, underrepresented languages.

One strand of his research addresses Kokborok, a Tibeto-Burman language spoken mainly in Tripura, Northeast India. Concerned that the language lacks digital tooling available to more widely spoken languages, he built a KNN-based spell-checking and error-correction system for Kokborok, reaching 98% accuracy, and is now try to extending this into a full NLP resource: a curated corpus alongside a browser-based, real-time spell-check editor.

On the medical imaging side, he designed a disagreement-aware, multi-expert deep learning framework for breast cancer detection from ultrasound images. Rather than a single classifier, the system uses three specialized expert heads to independently assess texture, shape, and boundary features, then quantifies where they disagree to flag diagnostically ambiguous cases for expert review. Built on a single EfficientNetB3 backbone, it reaches 94.2% accuracy and a 0.952 AUC-ROC on the BUSI dataset while using roughly 66% fewer parameters than typical ensemble approaches.

He holds an M.Tech in Computer Science from NIT Silchar and a B.Tech in Computer Science and Engineering from Tripura Institute of Technology, and is comfortable across the full ML/DL stack: Python, TensorFlow, PyTorch, and Flask for deployment. This breast cancer detection framework was published and presented at the IEEE International Conference on AI Engineering and Innovation, NIT Jamshedpur, in March 2026.

Before moving into academia, he interned at the National Informatics Centre, building a BMI-tracking web application, and at Ardent, where he worked on weather-prediction models using classical machine learning methods. He is now looking to build out both of his ongoing research threads, medical AI and NLP for underrepresented languages, into a sustained research programme at DSU.

Email: This email address is being protected from spambots. You need JavaScript enabled to view it.

LinkedIn: https://www.linkedin.com/in/harijit-tripura/

GoogleScholar: https://scholar.google.com/citations?hl=en&user=xhDEMOgAAAAJ