matloptlib

Interactive Python app for graphing functions and exploring their derivatives and integrals, with built-in analysis tools.

Classification data and using ANN model

Synthetic loan portfolio dataset for Turkey, Germany, and Jordan. Built with Python and pandas — demonstrates relational data modelling, table merging (LEFT JOIN), Excel I/O, and BI visualization. Frankfurt School MENA Internship project.

A multivariate machine learning framework for predicting student stress levels with 90%+ accuracy.

End-to-end Sales Revenue Optimization project using SQL (MySQL), Python (Pandas + Matplotlib), and Power BI. Includes data cleaning, KPI generation, EDA, and an interactive business dashboard for retail performance analysis.

“Petshop platform with vet services, interactive React frontend, and Python data analysis for sales and pet records.”

Amazon Customer Review Sentiment Analysis using Python, Pandas, TextBlob and Matplotlib.

Data cleaning and exploratory analysis of 186K retail electronics transactions (2019) with pandas & matplotlib — answers 5 business questions in a fully reproducible notebook.

🔵 School Project - Epitech - 2nd year - Zoidberg 2.0 est un système d'aide au diagnostic médical (Computer Aided Diagnosis) capable de détecter une pneumonie sur une radiographie pulmonaire, construit en machine learning classique (scikit-learn) sur un dataset de 5 856 radios thoraciques.

Python web scraping project using YouTube API to analyze popular data science educators, comparing their channel performance and video content patterns.

HR, ESG & Finance Analytics Dashboard – headcount, salaries, attrition, and trends (synthetic 2024 data). Built with Power BI and Python.

Real-time ergonomic posture risk analysis using MediaPipe pose estimation and computer vision.

Business case study - Data Science and machine learning

Path planning visualizer. A* & BFS on randomized grids, step-by-step GIF export.

Developed a machine learning model to predict Uber ride prices. Evaluated multiple regression models, including Linear Regression, Ridge, Lasso, ElasticNet, and Random Forest. Conducted comparative analysis to optimize model selection for accuracy and interpretability. Identified key pricing determinants such as distance and time-of-day effects

E-commerce customer retention, revenue concentration and delivery analysis using Python and DuckDB

Heuristic Battleship solver using probability density mapping to determine optimal firing sequences

Simple Linear Regression that predicts the GPA based of students using OLS.

Statistical analysis of bike rental demand in Python (pandas, matplotlib, seaborn) examining time-of-day and weather drivers