Release Time:2019-03-11  Hits:

Indexed by: Conference Paper

Date of Publication: 2014-08-08

Included Journals: Scopus、EI

Page Number: 237-242

Abstract: An attribute weighted fuzzy c-means algorithm for incomplete datasets is presented in this paper. The nearest-neighbor interval representation proposed in our previous researches is used here to describe the missing attribute values, and the widely used ReliefF algorithm is involved to determine the attribute weights. The proposed algorithm combines the attribute weights and fuzzy clustering by weighted Euclidean distance, and is successfully justified based on benchmark datasets. ? 2015 Taylor & Francis Group, London.