Classification using information statistics.

R. J. Johnston, R. K. Semple

Research output: Contribution to journalArticlepeer-review

26 Citations (Scopus)

Abstract

Information theory provides a means of analysing closed number sets, to categorize an individual distribution or to compare distributions. An inequality statistic, L, is derived which can be interpreted as a measure of the inequality of the distribution across components, and which is a special case of the general measure of information. A grouping procedure is introduced, which uses this statistic, so that all members of each group have similar distributions. An algorithm and FORTRAN computer program is provided to perform the procedure, which is illustrated by the application to social areas in cities, time series analysis, and trade flows. -A.C.Armstrong

Original languageEnglish
JournalCATMOG (Concepts & Techniques in Modern Geography)
Volume37
Publication statusPublished - 1983

Fingerprint

Dive into the research topics of 'Classification using information statistics.'. Together they form a unique fingerprint.

Cite this