Powerset, Inc., based in San Francisco, is on the verge of offering an innovative natural-language search engine, based on linguistic research at the Palo Alto Research Center (PARC). The engine does more than merely accept queries asked in the form of a question. The company claims that the engine finds the best answer by considering the meaning and context of the question and related Web pages.
“Powerset extracts deep concepts and relationships from the texts, and the users query and match them efficiently to deliver a better search,” Powerset CEO Barney Pell says.
Even though attempts have been made at natural-language search for decades, Powerset says that its system is different because it has solved some of the fundamental technological problems that have existed with this kind of search. It has done so by developing a product that is deep, computationally advanced, and still economically viable.
Pell says that it’s difficult to pinpoint one particular technological breakthrough, but he believes that Powerset’s superiority lies in the three decades of hard work by scientists at PARC. (PARC licensed much of its natural-language search technology to Powerset in February.) There was not one piece of technology that solved the problem, Pell says, but instead, it was the unification of many theories and fragments that pulled the project together.
“After 30 years, it’s finally reached a point where it can be brought into the world,” he says.
A key component of the search engine is a deep natural-language processing system that extracts the relationships between words; the system was developed from PARC’s Xerox Linguistic Environment (XLE) platform. The framework that this platform is based on, called Lexical Functional Grammar, enabled the team to write different grammar engines that help the search engine understand text. This includes a robust, broad-coverage grammar engine written by PARC. Pell also claims that the engine is better than others at dealing with ambiguity and determining the real meaning of a question or a sentence on a Web page. All these innovations make the system more adaptable, he says, so that it can extract deep relationships from text.
Powerset chief technology officer Ron Kaplan has led PARC’s XLE team since the 1970s and is the author of much of the technology behind XLE that has been licensed to the company. Kaplan says that he and Pell began to collaborate on the idea about two years ago.
Current methods of searching used by more traditional engines focus on isolated keywords and broad but shallow content coverage. This leaves a lot of room for improvement, Kaplan says.
“They are really not getting at relationships,” he notes. “The best that they do to approximate relationships are words that are close to other words.” He adds that a much deeper level of analysis is required.
Previous attempts have tried to pair some natural-language query processing with standard keyword searches of relevant content. This approach can be seen with some parts of standard search engines like Google, which, if it doesn’t understand a user’s query, will suggest another phrase or word that it thinks he or she may have meant. Engines such as Google and Yahoo use some components of natural-language search, yet there has not yet been a full-scale natural-language search engine for consumers. (See “The Future of Search.”) Pell says that this was mainly because the necessary technology was simply not ready. Engines that use natural-language components for aspects of the search, such as iPhrase and EasyAsk, don’t process textual content as Powerset does, Pell says, but instead simply query databases for answers to questions. Attempts at full natural-language search, such as that offered by Hakia and Cognition Search, do not cover as rich a representation of concepts or meaning, Pell says.
Hear more from Google at EmTech 2014.