Context-dependent fuzzy queries in SQLf

Abstract

Fuzzy set theory has been used for extending the database language capabilities in order to admit vague queries. SQLf language is one of the most recognized efforts regarding this tendency, but it has limitations to interpret context-dependent vague terms. These terms are used as filtering criteria to retrieve database objects. This paper presents an improvement of SQLf language that adding inductive capabilities to the querying engine. This allows the discovering of the semantics of vague terms, in an autonomous and dynamic way. In the discovering of the meaning of vague terms, looking for more flexibility, our proposal considers different granularity levels in the fuzzy partitions required for the object categorization. © 2012 Springer-Verlag.