Comparative Evaluation of Machine Learning Classifiers for Predictive Modelling of Datasets

This project deliverable contains the complete set of files associated with the project “Comparative Evaluation of Machine Learning Classifiers for Predictive Modelling of Datasets.”It provides all resources necessary to fully reproduce the analyses, preprocessing steps, model training procedures, and evaluation workflows conducted in the study. The upload includes: • Raw datasetsOriginal Wine and Colleges datasets used for classification experiments. • Preprocessed datasetsCleaned, normalized, and feature-engineered versions of the datasets used during training and evaluation. • Python source codeAll scripts used for data preprocessing, model training, performance comparison, and visualization.This includes implementations of:– Decision Tree Classifier– Gaussian Process Classifier (GPC)– Multi-Layer Perceptron (MLP)– Utility modules used for dataset loading and preprocessing– Plotting scripts used to generate comparative visualizations. • Trained models and resultsModel outputs, performance metrics, and comparison plots generated during the analysis. • MetadataA structured metadata file (metadata.json) and a detailed README describing file structure and workflow. • Final Data Management Plan (DMP)The completed DMP associated with this project, outlining data handling, documentation, storage, sharing, and preservation practices. Purpose of the Project The aim of this project is to perform a comparative evaluation of different machine learning classifiers on two widely used datasets. The study analyzes classifier performance with respect to accuracy, training behaviour, generalization ability, and robustness.The materials included in this record support reproducibility, transparency, and educational reuse. Intended Use The uploaded files are provided for: research reproducibility, teaching and learning in data science and machine learning, demonstration of data management best practices, and open educational use under the selected license.