Matthew Brotherton's repositories
2020-Heart-Failure-Prediction
This repository contains a notebook that examines the performance of various classification models on the Kaggle dataset: https://www.kaggle.com/datasets/andrewmvd/heart-failure-clinical-data. The best performing model was a Random Forest Classifier with 86.67% accuracy.
2024-4-Connected-StellarGraph-GCN-For-Malimg-Dataset
This repository implements a StellarGraph-based GCN approach that interprets malware byteplot RGB pixel values as nodes, and node differences are treated as edge weights. Using 4-connectedness, this method is evaluated on the Malimg dataset with 10-fold cross validation and a GCN hyperparameter grid search.
2021-Dengue-Fever-Prediction-Time-Series-
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).
2023-An-Analysis-of-Image-based-Malware-Classification-Using-Deep-Learning-Techniques
This repository contains a variety of CNN-based approaches (CNN, CNN-SVM, InceptionV3) used to classify families of malware in the Malimg, MaleX, and MMCC datasets under a variety of dataset compositions.