Evaluation of Functional Data Models for Database Design and Use
Kulkarni, Krishnarao Gururao
MetadataShow full item record
The problems of design, operation, and maintenance of databases using the three most popular database management systems (Hierarchical, CQDASYL/DBTG, and Relational) are well known. Users wishing to use these systems have to make conscious and often complex mappings between the real-world structures and the data structuring options (data models) provided by these systems. In addition, much of the semantics associated with the data either does not get expressed at all or gets embedded procedurally in application programs in an ad-hoc way. In recent years, a large number of data models (called semantic data models) have been proposed with the aim of simplifying database design and use. However, the lack of usable implementations of these proposals has so far inhibited the widespread use of these concepts. The present work reports on an effort to evaluate and extend one such semantic model by means of an implementation. It is based on the functional data model proposed earlier by Shipman[SHIP81). We call this 'Extended Functional Data Model' (EFDM). EFDM, like Shipman's proposals, is a marriage of three of the advanced modelling concepts found in both database and artificial intelligence research: the concept of entity to represent an object in the real world, the concept of type hierarchy among entity types, and the concept of derived data for modelling procedural knowledge. The functional notation of the model lends itself to high level data manipulation languages. The data selection in these languages is expressed simply as function application. Further, the functional approach makes it possible to incorporate general purpose computation facilities in the data languages without having to embed them in procedural languages. In addition to providing the usual database facilities, the implementation also provides a mechanism to specify multiple user views of the database.
Showing items related by title, author, creator and subject.
An Automatic Speech Recognition System Using Neural Networks and Linear Dynamic Models to Recover and Model Articulatory Traces Frankel, Joe; Richmond, Korin; King, Simon; Taylor, Paul (International Speech Communication Association, 2000-10)We describe a speech recognition system which uses articulatory parameters as basic features and phone-dependent linear dynamic models. The system first estimates articulatory trajectories from the speech signal. Estimations ...
Wester, Mirjam (Utrecht Institute of Linguistics OTS, 1998)In this paper, two methods for automatically classifying voice quality are compared: regression analysis and hidden Markov models (HMMs). The findings of this research show that HMMs can be used to classify voice quality. ...
Wright, Helen; Taylor, Paul A (1997)A method is introduced for using hidden Markov models (HMMs) to model intonational structure. HMMs are probabilistic and can capture the variability in structure which previous finite state network models lack. We show ...