Expanding our map of Stanford research

The release brings our content to 350,000 nodes comprised of 1.4 million entities and 5.5 million attributes, made possible by latest advancement in our "Wankel process" data processing system. We have developed new streamlined techniques to analyze relationships and eliminate redundancies in the large data sets we get by scraping public academic websites. We have the engine now to broaden the horizon of our maps beyond Stanford, as requested by our users..

Our recent innovations," says Scott Van Duyne, company chief scientist and co-founder, "target the problem of merging and disambiguating entities and facts found in a variety of lightly-structured public sources such as published papers, patents, news, profile pages, public data stores and so forth, all this with the goal of deducing the trajectories of collaborative research teams and their outputs. Scraping multiple sources means better overall recall, but it also means coping with many inconsistencies and uncertainties; our highly iterative Wankel process just buzzes through the data, builds and rebuilds the grid, and cancels out the noise..

The tripling of data brings more complexity and density to the maps, se have updated our Visible Legacy Navigator™ website with smoother graphics, faster navigation, flexible filtering and more powerful search to help users explore the data more easily.

Also please visit the "TechFinder" Technology Transfer Portal offered by the Stanford Office of Technology Licensing (OTL).