Home // IMMM 2018, The Eighth International Conference on Advances in Information Mining and Management // View article

ECF-means – Ensemble Clustering Fuzzification Means

Authors:
Gaetano Zazzaro
Angelo Martone

Keywords: Clustering Optimization; Data Mining; Ensemble Clustering; Fuzzy Clustering; k-means; Weka.

Abstract:
This paper describes a clustering optimization algorithm for Data Mining, called Ensemble Clustering Fuzzification (ECF) means, which combines many different clustering results in ensemble, achieved by N different runs of a chosen algorithm, into a single final clustering configuration. Furthermore, ECF is a simple procedure to fuzzify a clustering algorithm because each point in the original dataset is assigned to each cluster with a degree of membership. Moreover, a novel clustering validation index, called Threshold Index (TI), is also here defined. The proposed approach is applied to the well-known k-means clustering algorithm by using its Weka implementation and an ad-hoc developed software application. Two case studies are also here reported; the first one in the meteorological domain and the second one concerns the famous Iris dataset. All the outcomes are compared with the results of the simple k-means algorithm against which ECF seems to provide more effective and usable final configurations.

Pages: 20 to 27

Copyright: Copyright (c) IARIA, 2018

Publication date: July 22, 2018

Published in: conference

ISSN: 2326-9332

ISBN: 978-1-61208-654-5

Location: Barcelona, Spain

Dates: from July 22, 2018 to July 26, 2018