Fuzzy Clustering of Interval Data Based on Wasserstein Distances

摘要

Fuzzy clustering has been widely applied to many management areas, such as communications, port plainning, civil aviation, environment and food category. Recently, the recording of interval data has become a common practice with the advances in databas technologies, while the fuzzy clustering method for interval data is poor. This paper introduces a new fuzzy clustering method for interval data based on Wasserstein distances. Compared with the traditional fuzzy-c means algorithms, this method employs distribution information in the interval data, and we show its advantages.