Premium
An analytical evaluation of NIAM'S grammar for conceptual schema diagrams
Author(s) -
Weber Ron,
Zhang Yanchun
Publication year - 1996
Publication title -
information systems journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.635
H-Index - 89
eISSN - 1365-2575
pISSN - 1350-1917
DOI - 10.1111/j.1365-2575.1996.tb00010.x
Subject(s) - construct (python library) , computer science , schema (genetic algorithms) , natural language processing , feature (linguistics) , rule based machine translation , artificial intelligence , grammar , linguistics , programming language , information retrieval , philosophy
. Nijssen's Information Analysis Method ( niam ) provides a powerful grammar for generating conceptual schema diagrams. We evaluate this grammar systematically using an ontological model proposed by Bunge, Wand and Weber. Our analysis shows niam 'Sgrammar has many desirable features. Nevertheless, we conclude it is deficient in three respects. First, we argue that construct overload occurs in niam 'Sgrammar. Multiple real‐world features are represented by the same grammatical construct in niam . As a result, we predict users of niam will sometimes be confused about which characteristics of the real world are being represented by niam 'Soverloaded grammatical constructs. Second, we argue that construct redundancy occurs in niam . Multiple grammatical constructs can be used to represent the same real‐world feature. As a result, we predict users of niam will choose to employ only one of the grammatical constructs provided to represent a particular real‐world feature. Alternatively, they will equivocate in their choice of a grammatical construct and thus undermine their effectiveness and efficiency in using niam . Third, we argue niam has construct deficit. Some ontological constructs have no corresponding grammatical construct in niam . As a result, we predict users of niam will employ other grammars to represent real‐world features that cannot be represented by niam 'Sgrammar. We predict that using multiple grammars to represent real‐world features, however, will evoke other problems associated with attaining consistent representations among the scripts generated via the grammars.