Extensive EDA of the IBM telco customer churn dataset, implemented various statistical hypotheses tests and Performed single-level Stacking Ensemble and tuned hyperparameters using Optuna.

Predicting Nobel Physics Prize winners. Final project for Harvard CS109a 2017 edition https://github.com/covuworie/a-2017.

This repository contains implementation and evaluation scripts for various pre-trained deep learning models applied to binary classification of cats and dogs using transfer learning on a balanced dataset. Explore different architectures such as VGG16, VGG19, ResNet50, InceptionV3, DenseNet121, and MobileNetV2 fine-tuned for accurate classification.

Binary Classification using Machine Learning

This project aims to predict sepsis in patients using advanced machine learning models. The workflow encompasses data preprocessing, feature engineering, class imbalance handling, hyperparameter optimization, model training, evaluation, model card generation, and model registry management for reproducibility and scalability.

A self-challenged speedrun to best solve a customer loyalty binary classification problem with and without ML libraries

Get to know if you are Diabetic or NOT.

A machine learning project to classify breast cancer tumors as malignant or benign using the Wisconsin Breast Cancer Dataset. It compares multiple classification algorithms including Logistic Regression, KNN, SVM, Random Forest, and XGBoost.

A binary classification model, inspired by the "Titanic" Kaggle Challenge. Predicts whether or not a given passenger will survive, based on personal characteristics such as age, gender, and how much money their ticket cost.

A web application to see effect of C hyperparameter on classification boundary and marginal threshold in SVM.

This project focuses on predicting customer purchase behavior using machine learning models, with an emphasis on feature importance.

Binary Classification in R and application to classify patients with diabetes

Complete data analysis project for the Statistical learning course. Data from a store dataset coming from Kaggle is used.

In this exercise, the objective is to build a classifier only with the training data, with the goal of achieving the best performance possible on the validation data.

End-to-end ML project predicting term deposit subscriptions using bank marketing data.

A machine learning program in Java that makes binary classifications of images. The ML algorithm used is a variant of the Support Vector Machine algo, a simplified version that was designed to be more palatable for novice programming students. This program is being revised to be a multi-class machine learning program.

ML-based seed selection strategy for improving directed fuzzing by prioritizing security-relevant input samples.

predicting articles popularity using a combination of text mining and classification.