Akaike’s Information Criterion for Linearly Separable Clusters
Roberto N. Padua, Maria Eda B. Arado · Liceo Journal of Higher Education Research · 2013
Using the Akaike Information Criterion (AIC) in cluster analysis with linearly separable components, the paper demonstrates the superiority of using the vector of slopes as inputs to the K-Means algorithm over using the raw data in determining the number of clusters.  Keywords - AIC(Akaike’s Information Criterion), Kullback-Leibler information, cluster analysis, linear separability Â