Machine Learning Projects

Predictive modeling, regression, and classification systems engineered for accuracy and explainability.

House Rent Prediction
Regression • ANN

House Rent Prediction

A multi-layered Artificial Neural Network (ANN) optimized to predict rental prices based on complex feature sets including location, amenities, and market trends.

Keras TensorFlow Pandas Front-end: Streamlit Back-end: Python
Repo
Classification • RF

Wine Quality Analytics

Advanced classification system utilizing Random Forest and Gradient Boosting to predict wine quality grades with high precision across chemical composition data.

Scikit-learn XGBoost Seaborn Front-end: Streamlit Back-end: Python
Repo
Wine Quality
Breast Cancer
Healthcare • SVM

Breast Cancer Detection

Medical diagnostic tool implementing Support Vector Machines and Logistics Regression to classify tumors as malignant or benign with a focus on recall metrics.

MLflow Scikit-learn Feature Eng Front-end: MLflow UI Back-end: Python
Repo
Vision • Clustering

Glioma Tumor Grading

Sophisticated grading system for glioma tumors using radiomic features and unsupervised learning techniques to identify tumor stages and progression.

OpenCV PyTorch Medical AI Front-end: PyQt5 Back-end: Python
Repo
Glioma Tumor
Credit Risk
FinTech • XGBoost

Customer Risk Profiling

End-to-end risk analysis system for banking customers. Uses behavioral data to predict default probability and segment customers by risk level.

Imbalanced Data FastAPI XGBoost Front-end: FastAPI Swagger Back-end: Python
Repo
MLOps • Risk Analysis

Credit Risk Engine

Production-ready machine learning engine for financial risk assessment. Integrated with SHAP for explainability and Streamlit for real-time dashboarding.

Scikit-learn XGBoost MLflow FastAPI Front-end: Streamlit Back-end: FastAPI
Repo
Credit Risk Engine
Movie Recommender
Recommendation • NLP

Intelligent Movie Recommender

A personalized recommendation engine using hybrid filtering techniques and content-based analysis to suggest movies based on user preferences.

Scikit-learn Pandas Streamlit Front-end: Streamlit Back-end: Python
Repo
Unsupervised • Clustering

Customer Segmentation Analytics

Advanced clustering system using K-Means and PCA to identify distinct customer segments for targeted marketing strategies and behavioral analysis.

K-Means Scikit-learn Data Visualization Front-end: Seaborn Back-end: Python
Repo
Customer Segmentation