Imbalanced data sets are a special case for classification problem where the class distribution is not uniform among the classes. Typically, they are composed by two classes: The majority (negative) class and the minority (positive) class.

This repository has the code for implementation of Principal Component Analysis, Upsampling (SMOTE), Downsampling (Random Undersampler) and combined via SMOTETomek.

The Repository is created to cover undersampling and oversampling methods to deal imbalance problem.

Machine Learning aplicado al mantenimiento predictivo. Se realizaron 2 modelos: 1 por medio de clasificación binaria que predice si una máquina fresadora estará en riesgo de fallar o no, y el 2 modelo a través de clasificación multiclase que predecirá el modo de falla

Predicting whether Insurance claims will accepted or rejected for online Travel Agencies

Personal GitHub to host and shares my academic mini-projects related to my master degree.

Use Random Forest to prepare a model on fraud data. Treating those who have taxable income <= 30000 as "Risky" and others are "Good" and A cloth manufacturing company is interested to know about the segment or attributes causes high sale.

Proyecto final del curso de DataScience en CoderHouse que intenta ayudar a una entidad bancaria a la hora de decidir si emite una tarjeta de crédito al solicitante

Loan prediction using Random Forest, Decision tree, SMOTE and SMOTETOMEK techniques.

Develop Machine Learning Models to Predict the UCI Bank Telemarketing Dataset

Churn prediction means detecting which customers are likely to leave a service or to cancel a subscription to a service.

Classification of an imbalanced dataset using SMOTE oversampling technique and ML Algorithms - KNN , XGBoost and Naive Bayes classifier

Identifying rare event.

Предсказание оттока клиентов из банка

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.

Our main objective is to determine if the person will be afflicted by a coronary heart condition or not, therefore we drew several insights from that dataset that helped us understand the weighting of each feature and how they are interrelated.

Data Science - Random Forest Work