International Journal of Electrical and Computer Engineering (IJECE) Vol. 10, No. 5, October 2020, pp. 5507~5513 ISSN: 2088-8708, DOI: 10.11591/ijece.v10i5.pp5507-5513  5507 Journal homepage: http://ijece.iaescore.com/index.php/IJECE Comparative study on machine learning algorithms for early fire forest detection system using geodata Zouiten Mohammed 1 , Chaaouan Hanae 2 , Setti Larbi 3 1 Department Laboratory EH3D, Faculty Polydisciplinary TAZA, University Sidi Mohamed ben Abdelah of FEZ, Morocco 2,3 Laboratory Sciences and Advanced Technologies, Faculty polydisciplinary of Larache, University Abdelmalek Essadi Tetouan, Morocco Article Info ABSTRACT Article history: Received Feb 27, 2020 Revised Apr 25, 2020 Accepted May 10, 2020 Forest fires have become a great risk for countries. To minimize their impact and prevent this phenomenon, scientific methods have emerged. Notably machine learning algorithms and decision-making Geographical Information Systems. Therefore, a competitive spatial prediction model for early fire forest detection system using geodata can be proposed. This model can help researchers to predict forest fires and identify risk zonas. System using machine learning algorithm on geodata will be able to notify in real time the interested parts and authorities by providing alerts and presenting on maps based on geographical treatments for more efficacity and analyzing of the situation. This research extends the application of machine learning algorithms for early fire forest prediction to detection and representation in geographical information system (GIS) maps. Keywords: Fire forest detection Machine learning Random forest Support vector machine Voronoi Copyright © 2020 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Zouiten Mohammed, Department of Geography FPT TAZA, University Sidi Mohammed Ben Abdelah of Fez, FPT BP 1223, TAZA, Morocco. Email: mohammed.zouiten@usmba.ac.ma 1. INTRODUCTION Forest fires have become a crisis factor in the world. To measure the impact of forest fires, countries touch the economic, human, environmental, social side factors. Several causes are at the root of this problem like ignorance of people who have contact with forests, global warming and natural factors. In particular, Portugal is an affected country by this kind of disaster [1]. Between 1980 and 2005, almost 2.7 million hectares of forest were destroyed. In particular, the fires of 2003 and 2005 which affected 4.6% and 3.1% of the territory and which were tragic [2]. Predicting this phenomenon is the solution to minimize the damage. As a result, human intervention alone is insufficient. Therefore, it is necessary to rely on technological tools [3]: satellites, topography drones and sensors. Each country chooses the appropriate method according to these means. Other means can also be used to measure non-stationary factors such as meteorology [4]. Portugal has 162 stations providing data to be analyzed in real time [5]. Forest weather structures provide numerical indices for preventing and warning probability of fire such as the Canadian Fire Weather Index (FWI) [6]. It is a system for indexes calculations based on: temperature, relative humidity, rain, etc.). This system is not only used in Canada but it has also been used in some European countries including Portugal [7]. recently these indices have become part of meteorological databases. They are subject study for different contexts, the main one being the extraction of knowledge from data based on the notions of datamining [8]. Certainly, these databases are very important but faced with their volumes are little exploited. Therefore, decision makers must not be satisfied with simple statistical analyses. For analysing and understanding of this phenomenon. Emerging machine learning methods will replace