sikit-learn

This repository explores the use of machine learning to predict heart attack risk factors. It analyzes a dataset of heart failure patients, examining various features like age, medical history, lab results, and the occurrence of death events.

smart waste identification with DeepLearning

Machine Learning project for Iris Flower Classification using Random Forest | CodeAlpha Data Science Internship

Finding Donors, CharityML, a Supervised Learning Machine Learning Project.

machine learning project for to identify the risk of the credit using best trained model

Crop recommendation using optimized XGBoost and Decision Tree with SHAP explainability

modeling on loan dataset

This project is a production-ready text classification system built using BERT. It takes raw text input (e.g., customer issues) and predicts the most relevant category along with a confidence score.

An end-to-end breast cancer tumor stage prediction system leveraging machine learning, clinical and genomic data, and a Flask-powered web interface to deliver accurate, real-time predictions.

Modelo de regresión lineal múltiple para pronosticar ventas mediante inversión publicitaria e intervalos de confianza.

Machine learning API for Iris classification built with FastAPI and scikit-learn.

Binary classification system to detect fraudulent credit card transactions using Decision Tree and SVM models with feature analysis and evaluation metrics.

Medical condition prediction using TensorFlow neural network classifies patient conditions from clinical data using NLP-based text encoding and deep learning.

AI-powered network switch monitoring system using SNMP and anomaly detection

Multi-role AI agent system — customer service, HR portal & owner dashboard — built with LangGraph, GPT-4o, RAG, and ML predictions. Arabic + English support.

End-to-end neural network fraud detection project using a Dense ANN, class imbalance handling, feature preprocessing, threshold optimization, and model evaluation with ROC-AUC, Precision, Recall, and F1-score

Machine learning model for predicting diabetes using medical data, demonstrating an end-to-end ML pipeline with training, evaluation, and model persistence.

Modeling on a timeseries dataset