THE DESIGN AND IMPLEMENTATION OF A NARRATIVE DATA REDUCTION SYSTEM,
Abstract
The design and implementation are described of a system for collecting and analyzing narrative textual data. The data base consisted of clinical reports on which several normalizing rules with respect to language and format were imposed. Machine processing procedures consist of two basic units: a file maintenance package that performs all operations necessary for the creation, duplication, and updating of the files, and a Data Analysis and Reduction System (DARES) that implements text reduction and data analysis languages and the operating environment for them. The two languages have similar structures but differ in their interpretation. The basic compiler consists of three interpreters, the parser that performs syntactic and lexical analysis of the source language, a generator that defines rules by which source elements are translated to equivalent object forms, and a 'messager' that operates on erroneous statements in the source string. The prototype system was implemented on a CDC 160A computer. The compiler processes source strings at about 280 characters/sec; the interpreters require 2000 words of storage, the remaining 6000 words being used for definition and symbol tables. Sample specifications and outputs are given. In this typical example, about 450 pages of text were processed against 14 questions in slightly less than 35 minutes. Aspects of use of the system by persons without technical computer backgrounds are discussed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1966
- Accession Number
- AD0644205
Entities
People
- Andrew J. Singer
Organizations
- Case Western Reserve University