Searching for a compromise between satisfaction and diversity in database fuzzy querying
Olivier Pivert, Allel Hadjali, Gregory J. Smits · 2011
This paper deals with fuzzy queries and describes an approach that aims at providing users with a set of answers which satisfies a diversity criterion on one or several attributes.Different cases are considered and two types of algorithms are described.The first one, which has a linear complexity in terms of the number of tuples in the result, is suited to the case where the notion of similarity underlying the definition of diversity is crisp.The second one, based on a trial and error strategy, makes it possible to deal with fuzzy similarity, but its high complexity means that it can be employed only when a relatively small sets of tuples is used to increase diversity.