A data model for use with formatted and textual data

✍ Scribed by Desai, B. C. ;Goyal, P. ;Sadri, F.

• Publisher: John Wiley and Sons
• Year: 1986
• Tongue: English
• Weight: 786 KB
• Volume: 37
• Category: Article
• ISSN: 0002-8231
• DOI: 10.1002/(sici)1097-4571(198605)37:3<158::aid-asi7>3.0.co;2-a

✦ Synopsis

individualized data structures and operations have been devised for the various application areas of textual data. Consequently, their ability to integrate with other applications by sharing their data objects has been lost. In this article we present a data model based on the text data model, universal relation model and nonfirst nor. mal form relations, and discuss the applicability of this data model to text processing and document retrieval systems. I. Introduction Ordinarily, systems for managing formatted and unformatted data have developed independently of each other. Classical database management systems (DBMS) were developed to handle formatted data. A DBMS offers a data model to structure and maintain information in a precise way. In the relational model [S] the information is structured in relation with atomic values as their entries, whereas document processing and retrieval systems handle what is commonly called unformatted data. A number of attempts have been made to bridge the gap between conventional relational DBMS technology and document-processing systems [3,7,8,12,19,20,21]. These attempts have mostly centered around the idea of capturing the unformatted information within the relational data structures and the piecemeal extension of the relational query languages to provide some of the facilities of present-day information retrieval systems (IRS) and document/text processing systems. Schek [20] and Jaeschke and Schek [ 131 give examples of a simple IRS query and its rather complex formulation in conventional SQL. In this article, we investigate the feasibility of integrating the text data model, nonfirst normal form relation schemes and the universal relation model for use in information retrieval systems. Unlike formatted data, textual