Avoiding duplicate records in a database using a linguistic quantifier based aggregation - A practical approach

We show how Zadeh’s calculus of linguistically quantified propositions can be applied to avoid duplicate names in a publication database of a research institute. This problem, in its most general form, is studied in the literature by various communities. Here we focus on a specific scenario in which a need for its solution arises. Moreover, we make an attempt to apply fuzzy logic based concepts to solve it. The approach proposed yields promising results for the data for which it has been initially conceived and seems to be applicable also in a more general context. Its primary advantage is the ease and intuitiveness of customization. The main parameters take the form of linguistic quantifiers whose meaning is arguably familiar for an average user and which are fairly easy to be tuned to the changing requirements.