Hybrid clustering algorithm based on ISFLA and PFCM with application to UBSS

To improve the sensitivity to initial values, poor robustness, and easy to fall into local extreme values in traditional fuzzy clustering algorithms, a hybrid clustering algorithm coming to the improved Shuffled Frog Leaping Algorithm (SFLA) and Possibility fuzzy C-means (PFCM) clustering algorithm...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Chinese Control and Decision Conference S. 462 - 468
Hauptverfasser: Kui, Xia, Wei, Li, Jing, Wang, Xinrui, Hong
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 15.08.2022
Schlagworte:
Blind source separation
Clustering algorithms
Estimation
Gaussian random walk
Improved Shuffled Frog Leaping Algorithm
Learning systems
Mixed matrix estimation
Probability fuzzy C-mean
Robustness
Simulation
Sociology
Underdetermined Blind Source Separation
ISSN:1948-9447
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To improve the sensitivity to initial values, poor robustness, and easy to fall into local extreme values in traditional fuzzy clustering algorithms, a hybrid clustering algorithm coming to the improved Shuffled Frog Leaping Algorithm (SFLA) and Possibility fuzzy C-means (PFCM) clustering algorithm was proposed and applied to the problem of underdetermined blind source separation. The algorithm uses the optimization process of SFLA to replace the iterative process of PFCM's gradient descent method. The improved SFLA initializes the population through the current optimal reverse learning mechanism and adds a Gaussian random walk strategy into the local search of subgroups, which effectively improves the optimization ability of the algorithm. The simulation results show that the ISFLA algorithm has a better optimization effect compared with the traditional Shuffled Frog Leaping Algorithm and the Particle Swarm Optimization algorithm. Meanwhile, the algorithm after fusion improves the robustness, clustering accuracy, and searching ability of the fuzzy clustering algorithm, and successfully realizes the estimation of the underdetermined mixed matrix. The estimation accuracy and stability of the proposed algorithm are high.
ISSN:1948-9447
DOI:10.1109/CCDC55256.2022.10034348