Using Fuzzy Linear Regression to Estimate Relationship between Forest Fires and Meteorological Conditions

Authors

Hande G. Akdemir, Giresun University
Fatma Tiryaki, Yildiz Technical University

Abstract

Each year, millions of hectares of forest land are destroyed by fires causing great financial loss and ecological damage. In this paper, our aim is to study the effect of the variation of meteorological conditions on the total burned area in hectares, by using fuzzy linear regression analysis based on Tanaka’s approaches. The total burned area is considered a dependent variable. Air temperature (in ºC), relative humidity (in %), wind speed (in km/h) and rainfall (in mm/m² ) are considered to be independent variables. The relationship between input and output data is estimated using data provided in data mining literature. In our study, we apply fuzzy regression, using crisp/fuzzy input data and fuzzy output data expressed in linguistic terms.

Recommended Citation

Akdemir, Hande G. and Tiryaki, Fatma (2013). Using Fuzzy Linear Regression to Estimate Relationship between Forest Fires and Meteorological Conditions, Applications and Applied Mathematics: An International Journal (AAM), Vol. 8, Iss. 2, Article 21.
Available at: https://digitalcommons.pvamu.edu/aam/vol8/iss2/21