Effective business solutions

Textkernel translates text mining and information extraction into effective business solutions

Textkernel is developing new technology to produce advanced information extraction and text understanding engines, and to provide intelligent content management solutions for organisations with large volumes of textual data. Textkernel's main product is Textractor Enterprise, an adaptive multi-lingual Information Extraction engine. Textractor is easily trained on domain specific labeled examples to capture key information from documents, regardless of their language, format, layout or vocabulary. The result is an intelligent high accuracy conversion of unstructured text into easy-to-integrate matchable XML-structured output.

Motivation

In many types of business processes, unstructured text documents present a formidable challenge for the management of ever growing volumes of data and the effective use of incoming information streams.

Our approach

Research in Natural Language Processing has yielded a wealth of robust statistical pattern recognition methods based on machine learning. These methods easily and automatically acquire knowledge of linguistic patterns from small sets of examples. Thus it has become possible to build cost-effective and maintainable document understanding systems for any of the world's languages.

Structure from text

With Textkernel's advanced technology it is possible to automate any task that involves: extracting database records from text, converting flat unstructured text to marked-up content, classification of documents, scanning text for specific information patterns, standardization of terminology, and shallow text understanding. Structured information yields better insights, faster action, and more relevant matches to your information need.

Company information

Textkernel is a privately held commercial R&D spin-off from research in Natural Language Processing and Machine Learning of the ILK group at the University of Tilburg, the CNTS group at the University of Antwerp, and the Computational Linguistics group at the University of Amsterdam. Textkernel was founded in 2001 and has been profitable since 2003.

Background information

Download our corporate brochure for more information about Textkernel and our solutions for your business challenges. "Full power over textual data"

Read a good general level introduction to the state of the art in Information Extraction research for people not familiar with language technology and machine learning : "Information Extraction: Distilling Structured Data from Unstructured Text". Andrew McCallum. ACM Queue, volume 3, Number 9, November 2005.