This repository contains codes, datasets, results, and reports of a machine learning project on air quality prediction.

Testing out ClearML.

A credit risk text classification pipeline designed to simulate real-world modeling workflows. This project uses financial text data to predict borrower risk, incorporating data cleaning, NLP preprocessing, and model evaluation—emphasizing skills in feature engineering, model pipeline structuring, and explainable machine learning.

🗣️ Speech Type Detection is a Flask app to classifies text into categories like "Hate Speech," "Offensive Language," or "No Hate or Offensive Language" with 87.3% accuracy. It offers a user-friendly interface for text input and prediction, using machine learning algorithms. Idea for managing online inappropriate language. 🌐🔍.

random forest classification (with hyperparameter tuning) on heart disease dataset.

🔍 Evaluating Sampling Techniques for Healthcare Insurance Fraud Detection in Imbalanced Dataset.

A machine learning-based fake news detection system that classifies news articles as "FAKE" or "REAL" using Naive Bayes and Support Vector Machine (SVM) models. The project features a text preprocessing pipeline, model evaluation, and prediction capabilities, demonstrating practical accuracy and efficiency for real-world news verification.

A comprehensive framework for developing YOLO family models, featuring streamlined workflows for training, validation, testing, and deployment through easy-to-use config files, enabling flexible customization to suit various object detection tasks.