DATA RETRIEVAL BY COMPUTER. A CRITICAL SURVEY.
HEBREW UNIV JERUSALEM (ISRAEL)
Pagination or Media Count:
The report constitutes a critical survey of work in the field of data-retrieval from a theoretical viewpoint. The conditions to be fulfilled, ideally, by a data-retrieval system which distinguish it from other types of data-processing systems are enumerated. Section 1 deals with problems of sentence ambiguity, syntactic andor semantic, with special reference to context-dependence. Various fallacies and misconceptions are pointed out. Section 2 is devoted to the question of data-based versus text-based systems. Sections 3 and 4 discuss the ambivalent attitude to various theoretical results. Examples Questions of consistency, decidability, syntactic simplification, inadequate explication. Various attempts to overcome these difficulties are described, and their respective merits and defects discussed. Section 5 deals with the fallacious identification of conversation and cross -examination, which is traced back to a misunderstanding of a classical paper of Turing. Section 6 digresses, to discuss systems having a two-dimensional input. Section 7 points out the fallacy of indiscriminate application of considerations of efficiency, etc., which are in general not accompanied by careful analysis of the specific systems in question. The conclusion asserts the need to concentrate research on less ambitious lines than hitherto. Author
- Information Science
- Computer Programming and Software
- Computer Hardware