Adaptive information extraction from unstructured documents. (21st August 2007)

• Record Type: Journal Article
• Authors: Dezsenyi, Csaba; Dobrowiecki, Tadeusz P.; Meszaros, Tamas
• Is Part Of: International journal of intelligent information and database systems. Volume 1:Number 2(2007)
• Page Start: 156
• Page End: 180
• Publication Date: 2007-08-21
• Subjects: information extraction -- IE -- adaptive systems -- language engineering platform -- document analysis -- unstructured documents -- natural language documents -- information processing; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

The authors present a novel adaptive framework that enables efficient development of applications demanding complex document analysis. In processing natural language documents the task is to transform them into application specific structured form. Such transformation has to be designed taking into account various abstraction levels and granularity of the processing and the multitude of possibly related requests driving the application. The proposed solution is based on the adaptively planned and executed network of information processing modules. The paper presents an overview of the framework, with the focus on the adaptive mechanism. An illustrative pilot application is also provided.