QUIS: a query language capable of handling incomplete information and statistics

Abstract

The increasing number and use of database management systems calls for a adequate approach to the incomplete information problem. In this thesis we describe the problem and examine a number of solutions. A new query language, QUIS is defined. QUIS is a user friendly language based on Lipski's theories on incomplete information, the normal distribution and confidence intervals. A subset of the language was implemented to demonstrate its practicality.