scitkit-learn

An end-to-end Machine Learning web application utilizing FastAPI and Streamlit to classify Iris flower species based on physical measurements. Demonstrates a decoupled production architecture separating the core Scikit-Learn training pipeline, a low-latency REST API backend inference layer, and an interactive frontend dashboard user interface.

Data Science project analyzing 2024-2025 NBA stats to predict game performance using Linear Regression & OLS models.

Machine Learning project for detecting fraudulent credit card transactions using Logistic Regression and Random Forest.

Machine Learning solutions for a fintech. Improving the success rate of term deposits

A Sentiment Analysis project using Python and Natural Language Processing (NLP) to analyze text data, classify sentiments as positive, negative, or neutral, and generate meaningful insights from customer reviews.

An analysis of food deserts in the US using data science techniques

A comprehensive collection of machine learning projects demonstrating applications in classification, regression, and clustering across domains like finance, energy, and marketing.

Unsupervised k-Means clustering on Bundesliga 23/24 player data, discovering football positions from raw performance metrics

Exploratory Data Analysis (EDA) and data preprocessing on students' social media usage habits, academic impact, sleep patterns, mental health scores, and addiction levels using Python, Pandas, Seaborn, and Scikit-Learn.

Premier League match forecaster combining Elo ratings and a Poisson goals model, served through a FastAPI backend with a live prediction UI.

End-to-end NLP sentiment analysis of Amazon product reviews using TF-IDF, Logistic Regression, and Multinomial Naive Bayes (~80% accuracy).

Workshop de machine learning realizando na Reunião Secional de Ramos do IEEE de 2025 na Universidade Federal de Fortaleza

Unsupervised anomaly detection for X/Twitter accounts - a scikit-learn Isolation Forest over behavioral features (activity rates, ratios, account age), with SHAP explanations for every flag. Surfaces accounts for review, not bot detection.

Machine Learning powered phishing URL detection system with Flask backend and Chrome extension for real-time website protection.