Projects

Planet is a robust event planner application built with Next.js, React, MongoDB, and Redux, designed to streamline the planning and management of both personal and professional events. Featuring an intuitive interface and powerful backend functionalities, including secure user and admin roles management with Passport.js, Planet offers a comprehensive solution for event organization.

This project focuses on recognizing facial expressions using advanced deep learning techniques and data preprocessing strategies, including transfer learning with pre-trained models like VGG16, ResNet50, MobileNetV2, and Xception, and employing data augmentation for improved accuracy.

This project aims to fine-tune pre-trained BERT and ALBERT models for Medical Named Entity Recognition (NER) tasks using the i2b2/n2c2 dataset. It includes code for data parsing, model fine-tuning, and evaluation, with models uploaded to Huggingface for public use.