Interactive dashboard for predicting prediabetes risk using machine learning and SHAP interpretability. Built for clarity, modular benchmarking, and clinical transparency. Includes manual input prediction, threshold-based classification, SHAP visualizations, and model comparison across classifiers.

Predict and prevent telecom customer churn with ML. XGBoost model achieves 84% accuracy, SHAP analysis provides explainability, and interactive Streamlit app delivers real-time predictions. Turn data into retention strategies.

A comprehensive health data analysis project using PySpark and TensorFlow for medical diagnosis and outcome prediction. Features large-scale data processing and interactive notebooks.

Benchmarking ML models for predicting DT Cancer recurrence with nested CV and explaining predictions with SHAP (XAI).

Fichiers généaraux pour le projet 7 d'openclassrooms

Multi-Scale CNN for EEG signal classification with SHAP-based explainability — combining deep learning and interpretable AI for healthcare and brain–computer interface applications.

A data science project that predicts employee attrition using real-world HR data. Tackles imbalanced datasets using SMOTE and Class Weights, explains predictions with SHAP and Permutation Importance, and evaluates performance with Precision-Recall curves. Built with business level relavence and accuracy.

AI-powered phishing detection and prevention system with explainable ML and SOC-style dashboard

Wine quality multi-class prediction neural net model implemented using pytorch with model exploration and explanation using shap.

Create a machine learning model to help an insurance company understand which claims are worth rejecting and the claims which should be accepted for reimbursement.

A Kaggle competition to predict future sales

Machine Learning Project || Predict Flight Prices || EDA & Feature Engineering & Model Training || SHAP - LIME Analysis 🧠|| Performance Metrics || Live Web Application

Model Training and Deployment Scripts for Predictive Maintenance.

AI-powered anomaly detection system for SCADA networks using LSTM and explainable AI.

We tested 3 distinct Deep Learning models (the LSTM-GRU hybrid model turned out to be the best) to forecast AAPL price movements, focusing on the prior 90 days. We additionally incorporated RSI and MACD to better mimic the actual price. To provide more insights, we combined SHAP with GPT-2, to find out the proper time to invest in these stocks.

End-to-end machine learning pipeline to predict heart attack risk using CDC BRFSS 2023 data. Includes data cleaning, XGBoost modeling, SHAP explainability, SageMaker deployment, MLflow tracking, automated retraining with Airflow, and GenAI-powered plain-language explanations.

A complete pipeline to detect fraudulent transactions using e-commerce and credit card data.

Python package and CLI for data cleaning, model training, and interpretable feature importance analysis using SHAP values. FIR helps you identify which features most influence your target variable and what value ranges are optimal for those features given a tabular dataset.

Classified: At Risk — HR analytics capstone using Python to forecast employee churn and inform retention strategy at Salifort Motors.

End-to-end explainable AI pipeline for medical classification using Random Forest and XGBoost with SHAP and LIME for global and local interpretability. Designed for transparent, trustworthy machine learning in healthcare and research applications.