Abstract
This paper describes an information retrieval system with the function to support user's use of the retrieved documents using the text-level structure of documents. The text-level structure of each document is described by the occurrence of typical functional components in the text. Automatic detection of the components has been attempted in previous works using surface-level language processing. The proposed system firstly utilizes the text structure to conduct high-precision searches of documents or passages by distinguishing the role or function each concept plays in the text. It also allows browsing or skimming of retrieved texts, creating summaries on-the-fly with various levels of condensation specified by the user. Moreover, the system can search and display any unit of a text such as a sentence, a paragraph or a chapter. Comparison of relevant passages in retrieved documents across multiple texts is helpful for users to examine, analyze, compare and integrate texts and...
Users
Please
log in to take part in the discussion (add own reviews or comments).