We are a small, innovative company, based in northern New England. We custom-tailor advanced search engines to the needs of clients in the legal, medical, academic and government sectors. We work directly with clients to integrate our software into existing enterprise systems and we conduct focused research, on behalf of our clients, in areas of mutual interest.

Our current research is focused on social network analysis, semi-supervised learning and computational graph theory, including the clustering of large datasets, data visualization and electronic discovery.

Contact us

Coburn & Cuadrado Incorporated
PO Box 972
Middlebury, Vermont 05753


John Cuadrado, Ph.D., Chief Scientist

Aaron Coburn, Chief Technology Officer

Laurel Coburn, Chief Operations Officer


Coburn & Cuadrado is a commercial outgrowth of an academic, research-based project, based at Middlebury College and the National Institute for Technology and Liberal Education.

This research into new and advanced techniques for doing search began in 2001 with a primary focus on Latent Semantic Indexing (LSI). The software was developed and used successfully in a wide array of domains, from newswire sources to literature and from bioinformatics to image data. The LSI algorithms, however, do not scale well, and so it became impractical to handle large collections of data on the computer systems available.

By 2003 LSI was abandoned in favor of a different core engine, this time based upon the mathematical model of Graph Theory. To a user the results are nearly equivalent to LSI, but the software is not affected by the scalability issues encountered before. This technique, as before, was used in a diverse set of domains and tested with many different languages, including Chinese, Russian and all Western European languages.

All of these techniques are well suited for data visualization, allowing a user to see and interact with the relationship among documents and concepts. These visualizations show how data changes over time: when certain concepts are introduced and how the arrangement of significant elements in the data evolves. These visualizations also help a user understand and navigate clusters of information, showing both relevance and relationship.

The lead researchers formed Coburn & Cuadrado in order both to further develop many of the ideas from the academic project and to pursue research in new areas, particularly computational graph theory and models for data discovery and data visualization.

Copyright 2010 Coburn & Cuadrado Incorporated