Featured Projects
Explore some of my recent AI and Machine Learning projects.
Spam Detection System
This project aims to help users identify phishing attempts, promotional spam, scam messages, fraudulent links, and other unwanted communications before interacting with them. By leveraging machine learning, the system provides fast, accurate, and reliable spam detection through a clean and user-friendly web interface.
SVM
Deep Learning
98%
Accuracy
97%
Precision
3
Models
Mental Health Detection System
Predicts students' mental health scores using Machine Learning by analyzing social media usage, stress levels, sleep patterns, physical activity, and academic behavior. The system delivers accurate, real-time predictions through a professional regression pipeline and interactive dashboard.
Extra Trees
91%
R2 score
0.91
CV Score
8
Models
California House Price Prediction
Developed an end-to-end Machine Learning model to predict house prices in California using demographic, geographic, and economic features from the California Housing dataset. The project involved comprehensive data preprocessing, EDA, feature engineering, model training, hyperparameter tuning, and an accurate regression model for real-world housing price prediction.
HistGradient Boosting Regressor
Regression
91%
R2 score
0.91
CV Score
8
Models
Insurance Prediction
Developed an end-to-end Machine Learning project for Insurance Cost Prediction that predicts an individual's medical insurance charges based on factors such as age, BMI, gender, number of children, smoking status, and region. The final model can be used to estimate insurance costs for new customer information, demonstrating how machine learning can support data-driven insurance pricing and cost estimation.
Gradient Boosting
Regression
0.90
R2 score
0.83
CV R²
Potato Leaf Disease System
The Potato Disease Detection System is an AI-powered image classification application developed to identify potato leaf conditions from images. The system uses deep learning and computer vision techniques to classify potato leaves into three categories: Early Blight, Late Blight, and Healthy.
EfficientNetB0
Deep Learning
+2
98.15%
Test Accuracy
97.03%
Macro F1-Score
97.98%
Avg. Confidence
Smartphone Addict Prediction
Smartphone Addiction Predictor is a machine learning-based web application that assesses a user's risk of smartphone addiction based on their daily smartphone usage and behavioral patterns. Users enter information such as screen time, social media usage, gaming hours, sleep duration, notifications, app usage, stress level, and work/study impact. The application processes these inputs and predicts a Low, Moderate, or High addiction-risk level.
XGBoost
Machine Learning
+2
96.8%
Accuracy
97.11%
ROC-AUC
96.2%
F1-Score