Operations and the Properties on Non-First-Normal-Form Relational Databases
Hiroshi Arisawa, Kunihiko Moriya, Takao Miura · 1983
RELATIONAL DATABASES In this paper, non first normal form relations (or NFRa) are discussed. First the authors define compoai t ion of tuplea to introduce NFR and discuss some properties. Then canonical forma of NFRa are defined using Vest” operations. This is optimal in the sense that every 1NF relation can be always transformed into canonical ones and canonical forma have the desired properties to some extent. Also we shall consider data dependency and its effect to NFRa. Finally we consider some algorithms for updating tuplea in NFRa with their complexity. 1. INTRODDCTION Moat of the recent works concerned with database systems aaaume theoretical background of relational data model Cl 1. One of the reasons is due to its mathematical foundation by which we can logically construct and manipulate information without paying attention to physical representation. However, several problems have been pointed out by aome researchers. Among them, the reasonableness of the first normal form (or 1NF) is sometimes discussed because it excludes compound value aeta from domains [5]. In advanced application processing, we could take more complicated value sets. Basically compound-value problems come from “data semantics”, it should be observed from data-model views. In this paper, we will not pursuit this problem but extend relational model using