Conceptual Information Extraction and Retrieval from Natural Language Input

Lisa F. Rau
1988 Open research Areas in Information Retrieval  
A b s tra c t Recent progress in understanding texts in constrained domains has increased the potential for storing texts in conceptusil format. Storing texts in conceptual format has a number of advantages. A conceptued format facilitates and encourages natural language interrogation. It cam support imprecise and inexact queries, conceptual informiation summa rization, and ultimately, document translation. In this paper, a method for extracting information from texts in constrained domains is
more » ... escribed. Information extracted is stored in a conceptual format, and can be retrieved with natured language queries. Due to the conceptual format of both the queries and the information, partial or inexact matches of desired information is supported through the representation of concepts in the domain. Information may also be summarized using heuristics for expository text summarization. These capabilities have been implemented in a prototype intelhgent information retrieval system called SCISOR (System for Conceptual In formation Summarization, Organization and Retrieval). In tro d u ctio n Typically, information obtained from natural lemguage texts, in the end, is ex tracted by human beings. Although relevant texts may be found through a process
dblp:conf/riao/Rau88 fatcat:far2tys225dcbjzkvefifunjsi