Research

Published research

Investigations into semantic patterns and topological structures across complex domains.

Study

Market Topology

Markets generate dense networks of information. Topology reveals how this information organizes, clusters, and flows.

Study

Not all clustering algorithms handle semantic embeddings equally. Density-based methods consistently outperform centroid methods for text data.

•Density-based clustering (DBSCAN, HDBSCAN) outperforms k-means for semantic embeddings
•Semantic clusters have irregular shapes that centroid methods cannot capture
•Dimensionality reduction improves results; 10-50 dimensions is often optimal

Study

Financial markets generate complex, high-dimensional data. Topological data analysis reveals structural patterns invisible to traditional statistics.

•Topological methods reveal market structure invisible to traditional statistics
•Persistent homology identifies features that persist across time and scale
•Topology captures regime changes earlier than mean/variance measures