ReVerb: The Whole Language Movement
August 12, 2011
Reverb, a new search method, presents an optimistic future for search engines and intelligence levels. Projecting what Web search engines will look like in ten years, ReVerb should hope that the whole language movement doesn’t make a comeback in schools. Requiring users to input an “argument” and a “predicate,” this program automatically identifies and extracts binary relationships from English sentences—and requires users to know the basic parts of a sentence.
Created by the University of Washington’s Turing Center, as a part of the KnowItAll project, there are currently 15 million Reverb extractions available for academic use. This program has blown similar ones out of the water.
The paper entitled, “Identifying Relations for Open Information Extraction” asserts the following:
“[ReVerb] more than doubles the area under the precision-recall curve relative to previous extractors such as TextRunner and WOE-pos. More than 30% of ReVerb’s extractions are at precision 0.8 or higher— compared to virtually none for earlier systems.”
The creators are confident that ReVerb will be useful for queries where target relations cannot be specified in advance and speed is important. Currently, there is a demo available.
Is this the next big thing in search or another public relations push? Will this generate sympathetic vibrations within the Google?
Megan Feil, August 11, 2011
Sponsored by Pandia.com, publishers of The New Landscape of Enterprise Search
Comments
-
- Subscribe to Beyond Search
Feature archive
News archive
-
