Investigation of general indicators influencing on forest fire and its susceptibility modeling using different data mining techniques

Pourtaghi, Zohre Sadat; Pourghasemi, Hamid Reza; Aretano, Roberta; Semeraro, Teodoro

doi:10.1016/j.ecolind.2015.12.030

Cited by 217 publications

(138 citation statements)

References 41 publications

Supporting

Mentioning

124

Contrasting

Unclassified

Order By: Relevance

“…The city had a total area of around 7029 km 2 and discussed in detail. The graphs of the likelihood of fire occurrence predicted by each model are also a direct demonstration.…”

Section: Study Areamentioning

confidence: 99%

“…According to previous studies, many of which were done in forest regions, human-related predictors are critical for explaining fire occurrence on the large scales, such as in Europe and China [2][3][4][5][6]. However, few studies were done at the city scale to explain and predict of the occurrence of infrastructure fire, which may lead to a lack of efficient management for the potential fire risks hidden in a city.…”

Section: Introductionmentioning

confidence: 99%

“…However, how fire occurs and spreads is highly complex and it is still difficult to explain the reasons and predict future incidents. Fire is similar to other natural and human disasters that are imbued with uncertainty and occur in dynamic systems with biologically diverse and complicated structures [2]. Using temporal and spatial datasets, along with historical datasets of fire ignition, it is possible to build valid and meaningful models for explaining fire occurrence; therefore, we can adopt these results to benefit the management of fire safety, from which we could assess the conditions of fire occurrence from a quantitative viewpoint.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, normalized differential vegetation index (NDVI), elevation, slope, aspect, and land use are popular factors used to assess fire risk hazard [2]. The parameters and variables used to train a model have a strong influence on how successful the model may be according to its statistical performance.…”

Section: Introductionmentioning

confidence: 99%

“…On the other hand, although most previous studies have offered several cases of fire occurrence modeling at the national scale using ML such as random forest (RF), whether ML can make robust and reliable predictions on fire occurrence at the city scale and whether there exist similar regularities about selected predictors at different space scales need to be examined further [2,10,17]. Therefore, we used RF as a comparison and discussion with SE models considering their advantages and excellent predictive ability.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Comparison between Spatial Econometric Models and Random Forest for Modeling Fire Occurrence

et al. 2017

View full text Add to dashboard Cite

Abstract:Fire occurrence, which is examined in terms of fire density (number of fire/km 2 ) in this paper, has a close correlation with multiple spatiotemporal factors that include environmental, physical, and other socioeconomic predictors. Spatial autocorrelation exists widely and should be considered seriously for modeling the occurrence of fire in urban areas. Therefore, spatial econometric models (SE) were employed for modeling fire occurrence accordingly. Moreover, Random Forest (RF), which can manage the nonlinear correlation between predictors and shows steady predictive ability, was adopted. The performance of RF and SE models is discussed. Based on historical fire records of Hefei City as a case study in China, the results indicate that SE models have better predictive ability and among which the spatial autocorrelation model (SAC) is the best. Road density influences fire occurrence the most for SAC, while network distance to fire stations is the most important predictor for RF; they are selected in both models. Semivariograms are employed to explore their abilities to explain the spatial structure of fire occurrence, and the result shows that SAC works much better than RF. We give a further explanation for the generation of residuals between fire density and the common predictors in both models. Therefore, decision makers can make use of our conclusions to manage fire safety at the city scale.

show abstract

“…The city had a total area of around 7029 km 2 and discussed in detail. The graphs of the likelihood of fire occurrence predicted by each model are also a direct demonstration.…”

Section: Study Areamentioning

confidence: 99%