Information Retrieval Research:
How It Might Affect the
Information retrieval as a discipline has been around for almost three decades. Also known as document retrieval, information retrieval, text matching, and text retrieval, information retrieval involves a user with an information need searching through a corpus or collection of documents, with each document containing predominantly textual information. The user expresses his/her information need as a query, perhaps a natural language query, and the "system" matches the user's query against each of the stored document texts and retrieves those texts most likely to be relevant to the user's original information need.
Traditionally, the main applications of information retrieval have been in libraries, patent offices, legal offices, and so on--almost anywhere where there is a large collection of objects of a textual nature among which users will wish to browse or search to satisfy some information need. In a library a user may wish to find a book about some topic, and in a legal environment a lawyer may wish to search among court cases for a precedent for a case he may be working on.
Recent advances in computer technology have meant that in many office environments, office documents such as letters and memos are now being generated and stored electronically and printed on high-quality laser printers. Even more recent technological advances mean that personal workstations with bitmapped screens, windows, mouse and menu interfaces, local magnetic storage of 100 megabytes or more, and local area connections like ETHERNET to file servers and print servers are starting to appear in the marketplace. It is possible to purchase a workstation with all of the above trappings and software for WYSIWYG (what you see is what you get) desktop publishing for a cost that is not beyond the range of many offices.