Wildfire spatial pattern analysis in the Zagros Mountains, Iran

A comparative study of decision tree based classifiers

Abolfazl Jaafari, Eric Zenner, Binh Thai Pham

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Naïve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8%, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7%, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.

Original languageEnglish (US)
Pages (from-to)200-211
Number of pages12
JournalEcological Informatics
Volume43
DOIs
StatePublished - Jan 1 2018

Fingerprint

Pattern Analysis
Spatial Analysis
Spatial Pattern
Decision trees
wildfires
wildfire
Decision tree
Comparative Study
Iran
comparative study
Classifiers
Classifier
mountains
mountain
Fires
fire suppression
Classification and Regression Trees
roughness
logit analysis
Wind effects

All Science Journal Classification (ASJC) codes

  • Ecology, Evolution, Behavior and Systematics
  • Ecology
  • Modeling and Simulation
  • Ecological Modeling
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

@article{5148311b4d34452aa59a2e5ab59860c4,
title = "Wildfire spatial pattern analysis in the Zagros Mountains, Iran: A comparative study of decision tree based classifiers",
abstract = "Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Na{\"i}ve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8{\%}, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7{\%}, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.",
author = "Abolfazl Jaafari and Eric Zenner and Pham, {Binh Thai}",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.ecoinf.2017.12.006",
language = "English (US)",
volume = "43",
pages = "200--211",
journal = "Ecological Informatics",
issn = "1574-9541",
publisher = "Elsevier",

}

Wildfire spatial pattern analysis in the Zagros Mountains, Iran : A comparative study of decision tree based classifiers. / Jaafari, Abolfazl; Zenner, Eric; Pham, Binh Thai.

In: Ecological Informatics, Vol. 43, 01.01.2018, p. 200-211.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Wildfire spatial pattern analysis in the Zagros Mountains, Iran

T2 - A comparative study of decision tree based classifiers

AU - Jaafari, Abolfazl

AU - Zenner, Eric

AU - Pham, Binh Thai

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Naïve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8%, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7%, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.

AB - Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Naïve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8%, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7%, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.

UR - http://www.scopus.com/inward/record.url?scp=85039173115&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039173115&partnerID=8YFLogxK

U2 - 10.1016/j.ecoinf.2017.12.006

DO - 10.1016/j.ecoinf.2017.12.006

M3 - Article

VL - 43

SP - 200

EP - 211

JO - Ecological Informatics

JF - Ecological Informatics

SN - 1574-9541

ER -