The Electronic Labyrinth

Full-Text and Information Retrieval

Full-text refers to unstructured, free, natural language text. This term is usually used in contrast to fixed-structure, record-oriented, or otherwise restricted text, such as that found in a DBMS. Indexing and other common database techniques are complicated in full-text, since every word is, by default, equal in importance to every other. The structure of the data cannot be as easily used to optimize indexing. Semantic processing is often necessary to produce useful navigation aids and search techniques.

Those working in the fields of library and information sciences view hypertext in a manner shaped by this dichotomy between full-text and database structures.

Software which falls under this heading includes Ize, AskSam, Folio VIEWS, and SmarText.

© 1993-2000 Christopher Keep, Tim McLaughlin, Robin Parmar.
contact us