In time for the 300th anniversary of Benjamin Franklin’s birth, a Web portal (http://ben.clusty.com) based on clustering technology is offering a new method to separate useful Franklin facts from the normal flood of online information.
Developed and curated by Vivísimo, Inc. of Pittsburgh, Pa., the portal uses the company’s unique clustering search engine. The underlying algorithms filter and index results to create clusters that are easier to navigate than numerous pages of “hits.”
Curators manage the portal, which is a comprehensive, searchable resource of the iconic statesman’s writings and quotations, along with a targeted search engine and tools for teachers. By adding the words “images” or “pictures” to a search, users can access related visuals.
The clustering component, known publicly by the name Clusty.com, works with any set of electronic documents, and as applications have arisen, developers have applied it to languages ranging from Arabic to Korean. The clustering is gaining wider recognition as it evolves, and the portal expands upon such evolution by highlighting the benefits of combining this automated technology with curation.
Franklin was an important figure not only in U.S. history, but also for science and engineering. From studies of electricity, weather and ocean currents to his development of the lightning rod, double spectacles (bifocals) and the odometer, many of his innovations and discoveries were groundbreaking.
A vast wealth of material on Ben Franklin exists on the Internet, but standard searches do not differentiate between references to the man and references to countless objects and entities named in his honor. The curators of the Web portal have simplified such searches by managing the results to weed out distracting references – from high schools to retailers – unrelated to Franklin himself.
Carnegie Mellon University computer scientist Raul Valdes-Perez and graduate students Chris Palmer and Jerome Pesenti created the underlying clustering algorithms in the early 1990s – with support from the National Science Foundation (NSF) and Pa.’s appropriately named Ben Franklin economic development initiative. The researchers founded Vivísimo and improved upon the technology, furthering the engine’s development with NSF Small Business Innovation Research grants.
Since its development, various users have applied the software to both institutional and World-Wide Web searches. Most recently, the U.S. General Services Administration chose Vivísimo’s platform for a re-launch of the FirstGov.gov search, the U.S. government’s official search gateway for government web pages.
NSF’s monetary support for Valdes-Perez’s work began more than a decade ago with funding from the Directorate for Computer & Information Science & Engineering and continued with grants from the Directorate for Engineering.
As Franklin himself has noted, “An investment in knowledge always pays the best interest.”
In addition to NSF support, the portal effort has been aided by the Pennsylvania Commonwealth Libraries, the Carnegie Science Center and Senator John Heinz Pittsburgh Regional History Center.
