Open AccessA Deterministic Seeding Approach for k-means Clustering
Author(s) -
Omar Kettani
Publication year - 2021
Publication title -
international journal of scientific research in computer science, engineering and information technology
Language(s) - English
Resource type - Journals
ISSN - 2456-3307
DOI - 10.32628/cseit217246
Subject(s) - cluster analysis , benchmark (surveying) , silhouette , seeding , computer science , algorithm , cure data clustering algorithm , correlation clustering , data mining , mathematics , artificial intelligence , geography , engineering , geodesy , aerospace engineering
In this work, a simple and efficient approach is proposed to initialize the k-means clustering algorithm. The complexity of this method is O(nk), where n is the number of data and k the number of clusters. Performance evaluation was done by applying this approach on various benchmark datasets and comparing with the related deterministic KKZ seed algorithm. Experimental results have demonstrated that this approach produces more consistent clustering results in term of average silhouette index.