INTEGRATING NATURAL-LANGUAGE UNDERSTANDING WITH DOCUMENT STRUCTURE-ANALYSIS

Document understanding, the interpretation of a document from its imag e form, is a technology area which benefits greatly from the integrati on of natural language processing with image processing. We have devel oped a prototype of an Intelligent Document Understanding System (IDUS ) which employs several technologies: image processing, optical charac ter recognition, document structure analysis and text understanding in a cooperative fashion. This paper discusses those areas of research d uring development of IDUS where we have found the most benefit from th e integration of natural language processing and image processing: doc ument structure analysis, optical character recognition (OCR) correcti on, and text analysis. We also discuss two applications which are supp orted by IDFUS: text retrieval and automatic generation of hypertext l inks.