Global intuitionistic fuzzy weighted C-ordered means clustering algorithm

The paper proposes a novel approach to address the challenges of clustering datasets and identifying outliers by utilizing the Atanassov intuitionistic fuzzy sets (AIFS) environment. The approach provides a more flexible and nuanced solution by incorporating a new function called the typicality func...

Full description

Saved in:
Bibliographic Details
Published in:Information sciences Vol. 642; p. 119087
Main Authors: Kaushal, Meenakshi, Garg, Harish, Lohani, Q.M. Danish
Format: Journal Article
Language:English
Published: Elsevier Inc 01.09.2023
Subjects:
Atanassov intuitionistic fuzzy set
Fuzzy clustering
Outlier detection
Atanassov intuitionistic fuzzy set
Outlier detection
Fuzzy clustering
ISSN:0020-0255, 1872-6291
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The paper proposes a novel approach to address the challenges of clustering datasets and identifying outliers by utilizing the Atanassov intuitionistic fuzzy sets (AIFS) environment. The approach provides a more flexible and nuanced solution by incorporating a new function called the typicality function for outlier detection and tuning the parameters used in defining it to improve clustering. To optimize the parameter tuning process and reduce time complexity, the paper introduces a global error search approach (k-GESA) within a finite space. The paper also presents a new clustering algorithm, called Global Intuitionistic Fuzzy Weighted C-Ordered Means (Global-IFWCOM), which utilizes k-GESA to improve the clustering results. The proposed approach is evaluated against various C-ordered means algorithms, including Fuzzy C-ordered means (FCOM), Intuitionistic fuzzy C-ordered means (IFCOM), Intuitionistic fuzzy weighted C-ordered means (IFWCOM), Fuzzy weighted C-ordered means (FWCOM), and Hesitant-FWCOM, on a two-dimensional synthetic dataset with outliers. Furthermore, the effectiveness of Global-IFWCOM is demonstrated using a six-dimensional synthetic dataset with noise and outliers compared to FWCOM. •Global error search approach (k-GESA) is proposed.•Developed a novel Weighted C-means clustering algorithm.•Performance evaluation with state-of-the art is done.•Two and six dimensional synthetic datasets are taken to demonstrate.
ISSN:0020-0255
1872-6291
DOI:10.1016/j.ins.2023.119087