My personal notes for mastering ML from theoretical derivation to code implementation.
NLP pipeline predicting BBC News categories from raw article text — Bag-of-Words features + Logistic Regression, with full evaluation and model coefficient interpretation.
Repositório com análises dos módulos de Estatística I e II da pós-graduação em Engenharia e Análise de Dados. O estudo usa o dataset Students Performance, com 1.000 alunos e dados pessoais, socioeconômicos e notas em Matemática, Leitura e Escrita. O notebook aplica probabilidade, Bayes, variáveis aleatórias e visualizações.
A data-driven restaurant decision intelligence platform using sentiment analysis, complaint detection and Tableau.
A Machine Learning project that predicts customer churn using Python, Pandas, and Scikit-learn. Includes data preprocessing, exploratory data analysis, classification models, and model evaluation.
TripAdvisor Travel Reviews verisinde 7 makine öğrenmesi algoritmasının (Karar Ağacı, Rastgele Orman, SVM, KNN, MLP vb.) çok ölçütlü karşılaştırması: çapraz doğrulama, etiket sızıntısı analizi ve K-Means segmentasyonu. Akademik makale dahil.
End-to-end Customer Churn Prediction using XGBoost, Scikit-Learn Pipeline, ColumnTransformer, SimpleImputer and RandomizedSearchCV.
Live poker hand equity calculator — Monte Carlo simulation + a trained ML model (Random Forest) for instant win-probability predictions. Runs entirely client-side (vanilla JS), no backend. Built with a Python data pipeline: simulation → feature engineering → model training → JS export.
Multi-class classification of student outcomes (Dropout / Enrolled / Graduate) on 4,424 students. 5 models are evaluated on Accuracy, AUC, Precision, Recall, F1 and MCC.
To associate your repository with the scilkit-learn topic, visit your repo's landing page and select "manage topics."