Information graphs for binary predictors

G Hughes, N McRoberts, FJ Burnett

Research output: Contribution to journalArticle

4 Citations (Scopus)
1 Downloads (Pure)

Abstract

Binary predictors are used in a wide range of crop protection decisionmaking applications. Such predictors provide a simple analytical apparatus for the formulation of evidence related to risk factors, for use in the process of Bayesian updating of probabilities of crop disease. For diagrammatic interpretation of diagnostic probabilities, the receiver operating characteristic is available. Here, we view binary predictors from the perspective of diagnostic information. After a brief introduction to the basic information theoretic concepts of entropy and expected mutual information, we use an example data set to provide diagrammatic interpretations of expected mutual information, relative entropy, information inaccuracy, information updating, and specific information. Our information graphs also illustrate correspondences between diagnostic information and diagnostic probabilities.
Original languageEnglish
Pages (from-to)9 - 17
Number of pages9
JournalPhytopathology
Volume105
Issue number1
DOIs
Publication statusFirst published - 2015

Fingerprint

Entropy
Crop Protection
Datasets

Bibliographical note

1023370

Keywords

  • Diagnosis
  • Disease management
  • Entropy
  • Information theory

Cite this

Hughes, G ; McRoberts, N ; Burnett, FJ. / Information graphs for binary predictors. In: Phytopathology. 2015 ; Vol. 105, No. 1. pp. 9 - 17.
@article{d9d4426a504748d58bfe38b454b044af,
title = "Information graphs for binary predictors",
abstract = "Binary predictors are used in a wide range of crop protection decisionmaking applications. Such predictors provide a simple analytical apparatus for the formulation of evidence related to risk factors, for use in the process of Bayesian updating of probabilities of crop disease. For diagrammatic interpretation of diagnostic probabilities, the receiver operating characteristic is available. Here, we view binary predictors from the perspective of diagnostic information. After a brief introduction to the basic information theoretic concepts of entropy and expected mutual information, we use an example data set to provide diagrammatic interpretations of expected mutual information, relative entropy, information inaccuracy, information updating, and specific information. Our information graphs also illustrate correspondences between diagnostic information and diagnostic probabilities.",
keywords = "Diagnosis, Disease management, Entropy, Information theory",
author = "G Hughes and N McRoberts and FJ Burnett",
note = "1023370",
year = "2015",
doi = "10.1094/PHYTO-02-14-0044-R",
language = "English",
volume = "105",
pages = "9 -- 17",
journal = "Phytopathology",
issn = "0031-949X",
publisher = "American Phytopathological Society",
number = "1",

}

Information graphs for binary predictors. / Hughes, G; McRoberts, N; Burnett, FJ.

In: Phytopathology, Vol. 105, No. 1, 2015, p. 9 - 17.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Information graphs for binary predictors

AU - Hughes, G

AU - McRoberts, N

AU - Burnett, FJ

N1 - 1023370

PY - 2015

Y1 - 2015

N2 - Binary predictors are used in a wide range of crop protection decisionmaking applications. Such predictors provide a simple analytical apparatus for the formulation of evidence related to risk factors, for use in the process of Bayesian updating of probabilities of crop disease. For diagrammatic interpretation of diagnostic probabilities, the receiver operating characteristic is available. Here, we view binary predictors from the perspective of diagnostic information. After a brief introduction to the basic information theoretic concepts of entropy and expected mutual information, we use an example data set to provide diagrammatic interpretations of expected mutual information, relative entropy, information inaccuracy, information updating, and specific information. Our information graphs also illustrate correspondences between diagnostic information and diagnostic probabilities.

AB - Binary predictors are used in a wide range of crop protection decisionmaking applications. Such predictors provide a simple analytical apparatus for the formulation of evidence related to risk factors, for use in the process of Bayesian updating of probabilities of crop disease. For diagrammatic interpretation of diagnostic probabilities, the receiver operating characteristic is available. Here, we view binary predictors from the perspective of diagnostic information. After a brief introduction to the basic information theoretic concepts of entropy and expected mutual information, we use an example data set to provide diagrammatic interpretations of expected mutual information, relative entropy, information inaccuracy, information updating, and specific information. Our information graphs also illustrate correspondences between diagnostic information and diagnostic probabilities.

KW - Diagnosis

KW - Disease management

KW - Entropy

KW - Information theory

U2 - 10.1094/PHYTO-02-14-0044-R

DO - 10.1094/PHYTO-02-14-0044-R

M3 - Article

VL - 105

SP - 9

EP - 17

JO - Phytopathology

JF - Phytopathology

SN - 0031-949X

IS - 1

ER -