VCoFWMVIFCM: An open-source code for viewpoint-based collaborative feature-weighted multi-view intuitionistic fuzzy clustering

We present VCoFWMVIFCM, an open-source Python implementation of a multi-view fuzzy clustering algorithm based on Intuitionistic Fuzzy c-Means (IFCM). The method integrates adaptive view, feature, and sample weighting to account for varying importance and reduce outlier effects. Local neighborhood in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Software impacts Jg. 23; S. 100743
Hauptverfasser: Golzari Oskouei, Amin, Samadi, Negin, Bouyer, Asgarali, Tanha, Jafar
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.03.2025
Schlagworte:
Feature weighting
Intuitionistic fuzzy c-means
Multi-view fuzzy c-means
Neighborhood information
Sample weighting
Multi-view fuzzy c-means
Sample weighting
Neighborhood information
Intuitionistic fuzzy c-means
Feature weighting
ISSN:2665-9638, 2665-9638
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present VCoFWMVIFCM, an open-source Python implementation of a multi-view fuzzy clustering algorithm based on Intuitionistic Fuzzy c-Means (IFCM). The method integrates adaptive view, feature, and sample weighting to account for varying importance and reduce outlier effects. Local neighborhood information enhances noise resistance, while a density-based initialization ensures stable centroid selection. These mechanisms collectively improve clustering robustness and accuracy for multi-view data. The modular implementation allows flexible execution and reproducibility, addressing real-world applications where multiple data perspectives exist. The code is publicly accessible on GitHub under the MIT license. •Provides open-source Python code for multi-view intuitionistic fuzzy clustering.•Implements adaptive weighting for views, features, and samples in clustering.•Includes neighborhood-aware clustering to handle noise and improve robustness.•Initial centroids are selected via a density-based strategy to ensure stability.•Code is modular, well-documented, and publicly available on GitHub for use.
ISSN:2665-9638
2665-9638
DOI:10.1016/j.simpa.2025.100743