Premium
Information measures
Author(s) -
Soofi Ehsan S.,
Zhao Huimin,
Nazareth Derek L.
Publication year - 2010
Publication title -
wiley interdisciplinary reviews: computational statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.693
H-Index - 38
eISSN - 1939-0068
pISSN - 1939-5108
DOI - 10.1002/wics.62
Subject(s) - cluster analysis , computer science , data mining , context (archaeology) , cauchy distribution , pareto principle , set (abstract data type) , multivariate statistics , statistics , mathematics , artificial intelligence , machine learning , paleontology , biology , programming language
This article presents an overview of the concept of information about random outcomes and measures that quantify information provided by probability distributions. We also provide a few examples and illustrate applications of the information measures in a computationally intensive context, namely cluster analysis. Information measures for the multivariate normal, Cauchy, and Pareto distributions are presented. Three clustering algorithms are proposed. The algorithms are used to cluster variables and observations in a data set. Copyright © 2010 John Wiley & Sons, Inc. This article is categorized under: Statistical and Graphical Methods of Data Analysis > Information Theoretic Methods