It’s been suggested that computing solutions have become a lot like electricity or the telephone for many of us. We expect all of them to work all the time, we experience the end result of a whole lot of effort put into making it work, and most of us don’t spend a lot of time thinking about how this stuff works or even why it does what it does. Keeping this in mind, Lextek has recently released a new tool that will serve as a new outlet or phone line for developers.
The Lextek Profiling Engine is a code library for developers who are adding document filtering, routing and classification to their applications. The profiling engine is capable of performing a high speed analysis on an incoming stream of documents, e-mail or news feeds and then accurately routing, filtering or classifying them based on a set of queries or taxonomies, according to Lextek.
According to Robert Perry, a senior research analyst for Internet based applications at the Yankee Group in Boston, tools like the Lextek Profiling Engine are becoming increasingly important for developers.
“If you go to the Web and look at the end result, being an application that presents appropriate information to the user, the core of that requires that you categorize and describe the information that you have in a way that the application can pick the right information,” he explained. “It’s very important to the whole process that you categorize and correctly identify the information and do it in logical fragments and chunks so that you can pick the right piece of information and give it to the right user.”
Lextek’s Profiling Engine includes a set of more than 40 query operators and is capable of handling large taxonomies. This allows developers and users to specify specific documents according to what fulfills their needs and what should be filtered.
Art Pollard, president of Lextek International in Provo, Utah, explained that these query operators are one of the features of this profiling engine that differentiates it from the competition.
“Some of these query operators are simple – standard, phrase, proximity,” he explained. “Others get into relevancy ranking and probabilistic judgements and fuzzy logic. It really provides a wide range of functions.”
Pollard continued, stating that another differentiator is the profiling engine’s capacity. “As far as we’re aware, we handle larger queries than anyone else. We don’t have any internal limits.”
According to John Lehman, CEO of Sageware Inc. in Mountain View, Calif., Lextek’s profiling engine was the right choice for his company because it met its very specific needs. “Sageware makes an automatic content tagger. Our strength is that we know how to define subjects so that a document can be read like how you and I would read it,” Lehman said about Lextek’s profiling engine, which is an embedded function that runs underneath Sageware’s engine. “We needed a feature set that not only allowed us to use our sophisticated categories, but would run fast enough and still be attractive from a price point perspective. All of those things were and are true about our relationship with Lextek.”
One feature that Lehman suggested could enhance the profiling engine is virtual zoning, however, he has been assured by the folks at Lextek that it’s on its way in a future release.
The profiling engine is able to run on a variety of operating systems including Windows, Linux, Solaris, MacOS/OS X and development environments including MSVC, Delphi and Visual Basic.
Lextek International can be reached at www.lextek.com.
