AI Research Enthusiast | Active in AI Research and Project
- Languages: Python, C++, HTML, CSS, JavaScript
- Frameworks: Pytorch, Tensorflow, React
- Databases: MySQL
- Tools: Git, GitHub, VS Code, Jupyter Notebook
- Neural Network experiments
- AI and Data Science project
- Machine Learning & AI applications
Here are some of the projects I’ve worked on:
- X-IDS: an explainable Intrusion Detection System for Network Security using Neural Autoencoders, Gradient Boosting, and T5-Small Text Generation.
- Match Triad Benchmark: A comparative benchmark of CSP, Genetic Algorithm, and Simulated Annealing for solving student-tutor matching based on preferences.
- COPPA Risk Classification: Developed a machine learning pipeline to predict COPPA violation probabilities in mobile apps using metadata, EDA, feature engineering, and calibrated ensemble models.
- Fraud Detection: Built an end-to-end pipeline to classify water quality data reliability by cleaning raw measurements, engineering domain-specific features, and training an XGBoost model.
- Data Quality Classification: Designed a fraud detection system using Random Forest, LightGBM, and XGBoost models combined through weighted ensemble and threshold optimization for improved accuracy.
- Car Price Prediction: A car price prediction project that leverages structured automotive features and feature engineering to build an effective predictive model.