An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations

Renato Cordeiro De Amorim

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    6 Citations (Scopus)


    This paper presents an analysis of the number of iterations K-Means takes to converge under different initializations. We have experimented with seven initialization algorithms in a total of 37 real and synthetic datasets. We have found that hierarchical-based initializations tend to be most effective at reducing the number of iterations, especially a divisive algorithm using the Ward criterion when applied to real datasets
    Original languageEnglish
    Title of host publicationAdvances in Artificial Intelligence
    Subtitle of host publication11th Mexican International Conference on Artificial Intelligence - revised selected papers
    PublisherSpringer Nature
    ISBN (Electronic)978-3-642-37807-2
    ISBN (Print)978-3-642-37806-5
    Publication statusPublished - 2013
    Event11th Mexican Int Conf on Artificial Intelligence - MICAI 2012 - San Luis Potosi, Mexico
    Duration: 27 Oct 20124 Nov 2012

    Publication series

    NameLecture Notes in Computer Science


    Conference11th Mexican Int Conf on Artificial Intelligence - MICAI 2012
    CitySan Luis Potosi

    Cite this