Data Reduction and Feature Isolation for Computing Persistent Homology on High Dimensional Data

This peer-reviewed publication was produced as part of graduate research conducted at the University of Cincinnati College of Engineering and Applied Science.

The research focuses on identifying meaningful topological features within high-dimensional datasets while reducing computational overhead. The methods improve the practicality of persistent homology in complex analytical environments.

The results demonstrate how targeted reduction and feature isolation techniques can improve performance while preserving important structural characteristics of the data.