Date of Award
Doctor of Philosophy
Shashi K. Gadia
In temporal database systems the time varying aspects of data are captured by time-stamping data values. Research in temporal databases has concentrated on developing models in which it is essential that all the information be known;In the present work a relational model for incomplete information is presented. The model allows incomplete temporal information to be stored, and provides a powerful, yet simple, algebra to query the incomplete information;The incomplete information model presented here generalizes a well-known model for temporal databases with complete information. The algebraic expressions in the model produce results that are reliable in the sense that they never report incorrect information. This is shown by introducing the notion of completions of relations and databases. It is also shown that except for certain cases of selection, if the definition of the operators were strengthened to give more information, we could obtain results that are not reliable. This result is obtained by introducing the concepts of extensions of relations and more informative relations;Update operations create, change, and changekey are defined. These operations allow the user to modify the state of the database to reflect changes in the real world, to correct errors in the database, and to increase the information content of incomplete objects as more information becomes available.
Digital Repository @ Iowa State University, http://lib.dr.iastate.edu/
Sunil S. Nair
Nair, Sunil S., "A relational model for incomplete information in temporal databases " (1993). Retrospective Theses and Dissertations. 10175.