抄録
The need for dealing with imprecise data is more and more acknowledged in order to cope with real data, even if commercial systems are most of the time unable to manage them. This paper is concerned with the querying of possibilistic relational databases. In this framework, any imprecise piece of information is modeled as a possibility distribution intended for constraining the more or less acceptable values. Such an imprecise database has a natural interpretation in terms of a set of regular databases which provides the basic gateway to interpret queries. However, if this approach is sound it is not realistic and it is necessary to consider restricted queries for which a calculus grounded on the possibilistic database is feasible. Extended yes/no queries are dealt with here, i.e., yes/no queries in the presence of imprecise data. Their general form is: ""to what extent is it possible and certain that the answer to Q fulfills property P?"", where Q is an algebraic relational query. In this paper, property P concerns the membership (or not) to the result, of a set of tuples specified by the user.
