Open AccessPrediction of forest fire using ensemble method
Author(s) -
Dedi Rosadi,
Widyastuti Andriyani
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1918/4/042043
Subject(s) - boosting (machine learning) , adaboost , ensemble learning , decision tree , computer science , ensemble forecasting , artificial intelligence , support vector machine , cluster analysis , machine learning , data mining , computation , pattern recognition (psychology) , algorithm
In this paper we consider the application of ensemble classification method, which is called as the Adaptive Boosting (AdaBoost) method, to predict the occurrences of forest fire. To illustrate the method, we consider the application of the method using the same public data set, which has been used in the previous studies, but the ensemble approach is not considered in these studies yet. We also compare the performance of the ensemble method with several other classical classification methods, such as the Decision tree and SVM method. All computation are done using open source software R. We find that in the empirical study, the hybrid algorithms between the fuzzy c-means clustering and the ensemble approach will outperform the other classification methods considered in the study.