I enjoy making things. Here are a selection of projects that I have worked on over the years.
A data science project to predict the sale prices of bulldozers using historical sales data and machine learning techniques in Jupyter Notebook.
A comprehensive MLOps project implementing a CI/CD pipeline for deploying a sentence completion model using GPT-2 from Hugging Face, leveraging Azure and Docker for scalable, production-ready NLP solutions.
SkimLit is an advanced NLP-based project designed to classify sentences in biomedical abstracts into rhetorical categories like Objective, Methods, Results, Background, and Conclusion. It helps researchers quickly skim through large volumes of biomedical literature by automatically organizing abstract content.
A command-line tool for extracting meaningful noun phrases from Wikipedia articles using natural language processing techniques in Python.
A hybrid fake news detection system combining traditional machine learning with deep learning techniques. The project focuses on identifying misinformation by using both TF-IDF + Naive Bayes pipelines and Convolutional Neural Networks for text classification.
