Text-mining software developed at UQ

Text-mining software developed at The University of Queensland is attracting worldwide attention for its simple, easy-to-use and effective approach, not to mention its affordable price tag.
The Leximancer software allows single desktop users to analyse large amounts of unstructured text, such as a 100-page document or 10,000 emails, at a glance.

A pre-market release version of the software has been successfully trialled in security and intelligence, media monitoring, marketing, legal research, policy research and academic research by more than 100 clients in Australia, UK, Hong Kong, USA, Italy and Germany.

According to Leximancer’s creator Dr Andrew Smith, a researcher at the UQ Key Centre for Human Factors and Applied Cognitive Psychology, clients have consistently reported that the system is superior to existing text-mining solutions in its honest extraction and representation of meaning.

“We’ve demonstrated that of any text-mining software, Leximancer provided the most accurate reflection of the text’s meaning and this could be reproduced in multiple languages,” Dr Smith said.

“Critically, the advanced algorithms which underpin the software are not based on any human-imposed rules, allowing the text to drive the software to reveal systematic patterns inherent in its meaning.”

In developing Leximancer, Dr Smith eschewed the rules-based linguistics approach typically employed in text-mining and has instead used a complex systems approach drawing on Bayesian theory and machine learning.

The software is being commercialised by UQ’s main technology transfer company, UniQuest Pty Ltd.

UniQuest Managing Director David Henderson said because Leximancer ran contrary to established text-mining conventions, it was not surprising there had been some resistance from the market place.

“We’ve received strong interest from local and international intelligence and crime agencies as well as major companies looking for ways to manage marketing and consumer information,” Mr Henderson said.

“Sales of a pre-release version of Leximancer have been predominantly in the higher education and research market but we are seeking investors to find further development, including an enterprise version, and assist in bringing Leximancer to a much larger market.”

“In the long term, we believe the software has the potential to become a widely used knowledge-management tool by both individuals and corporates,” he said.

Link: UQ Key Centre for Human Factors and Applied Cognitive Psychology

Source: University of Queensland

Citation: Text-mining software developed at UQ (2005, June 15) retrieved 24 April 2024 from https://phys.org/news/2005-06-text-mining-software-uq.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

New report outlines opportunities to remove CO2 at the gigaton scale

0 shares

Feedback to editors