Klasifikasi kebakaran hutan menggunakan algoritma C4.5 dan Rough Set

Abstract

In recent years there have been large-scale forest fires in forested areas of the world. Forest fires are a major environmental problem that has big impact on wildlife, human health, economic. One solution can be taken is using classification algorithm to predict forest fires based on historical forest fire data.

In this research using C4.5 Algorithm combined with Rough Set as feature selection to classify forests fire. Evaluate performance based on created model using confusion matrix to calculate accuracy value.

The results show the C4.5 algorithm with Rough Set as feature selection was found accuracy 98.36%. The use of Rough Set as feature selection can reduce irrelevant attributes effectively.