You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
An ensemble of machine learning models for detecting fraudulent credit card transactions, utilizing advanced techniques for feature selection, data imbalance handling, and hyperparameter tuning.
This study aimed to assess whether machine learning algorithms would outperform traditional modeling in developing a cesarean delivery prediction model among gravidas with morbid obesity (body mass index of ≥40 kg/m2) to determine whether a primary cesarean delivery may be beneficial.
The research aims to harness machine learning for predicting cardiovascular diseases based on numerous risk factors, addressing the high fatality rates associated with cardiovascular conditions.
This repository contains the notebook used for the Spring 2021 Kaggle Dengue Fever Prediction Competition. Placement was in the top 10% with a MAE of 24.86. Our best approach involved Random Forest Regression on a reduced featureset selected with Recursive Feature Elimination in combination with correlation with the target (number of dengue cases).
Delved into advanced techniques to enhance ML performance during the uOttawa 2023 ML course. This repository offers Python implementations of Naïve Bayes (NB) and K-Nearest Neighbor (KNN) classifiers on the MCS dataset.
The goal of this project is to develop a predictive model that accurately detects failures in the Air Pressure System (APS) of heavy Scania trucks. The APS is responsible for generating pressurized air used in various functions of the truck, such as braking and gear changes. By detecting failures in the APS system.
Predict the attrition (Yes/No) of employees, identify factors significantly impacting it, and finally state recommendations on how to mitigate the attrition.
Used CDC dataset for heart attack detection in patients. Balanced the dataset using SMOTE and Borderline SMOTE and used feature selection and machine learning to create different models and compared them based on metrics such as F-1 score, ROC AUC, MCC, and accuracy.