Fuzzy Quantifiers: A Natural Language Technique for Data Fusion

Fuzzy quantifiers like ‘almost all’ and ‘about half’ abound in natural language. They are used for describing uncertain facts, quantitative relations and processes. An implementation of these quantifiers can provide expressive and easy-to-use operators for aggregation and data fusion, but also for steering the fusion process on a higher level through a safe transfer of expert-knowledge expressed in natural language. However, existing approaches to fuzzy quantification are linguistically inconsistent in many common and relevant situations. To overcome their deficiencies, we developed a new framework for fuzzy quantification, DFS. We first present the axioms of the theory, intended to formalize the notion of ‘linguistic adequacy’. We then argue that the models of the theory are plausible from a linguistic perspective. We present three computational models and discuss some of their properties. Finally we provide an application example based on image data.