Communications of the ACM
Computers Learn Better Reading Comprehension Through t Research
A researcher at the University of Twente has developed a way to help computers better understand written content.
Credit: Matanya Horowitz
University of Twente researcher Mena B. Habib has developed a way to help computers improve their understanding of written texts. Using context, the method enables a computer to analyze the importance of named entities such as people, places, and organizations.
Demand is rising for improved ways of helping computers comprehend text, and many current methods are based on superficial word analysis that leaves much of a text's meaning obscured. "With greater understanding of the entities referred to and information available about these entities, computers are better able to extract a lot more information from texts for analysis purposes," says University of Twente senior lecturer Maurice van Keulen.
The researchers are participating in the TEC4SE project, in which the software will be deployed in the emergency rooms of the Twente fire brigade and police. The software reads and understands tweets that can alert emergency services workers of problems. The method is language-independent and does not require a large number of texts from which to learn, van Keulen says.
From University of Twente (Netherlands)
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found