Classbound: Exploring Boundaries for Classifiers

This project aims to extend the classbound R package into a general framework for exploring and comparing classification boundaries across a wide range of models and datasets. The project will develop a modular R package that supports additional classification algorithms, including user-defined methods, along with flexible data simulation utilities and support for real datasets. It will also extend functionality to high-dimensional settings using tour-based visualization techniques, allowing users to interactively explore data structure and corresponding decision boundaries. The package will be designed for extensibility and will include comprehensive documentation, examples, and a vignette. Final deliverables will include implementations of additional classifiers, enhanced simulation capabilities, high-dimensional visualization support, and complete documentation, enabling users to better understand and compare classification models through their decision boundaries.