SUPPORT VECTOR MACHINES ALGORITHM FOR LAND COVER MAPPING USING GOOGLE EARTH ENGINE CASE STUDY SUB-WATERSHED OF OUED EL ARABE
Author(s): Chakali Ahmed Nadjib ,Zeraib Salah, Abderrahmane Ben Salem HACHI
J. Ponte - Dec 2020 - Volume 76 - Issue 12
doi: 10.21506/j.ponte.2020.12.10
Abstract:
With the rapid growth in big data, classification is becoming a necessary tool for
practitioners and decision-makers in many fields of research. Several classifications
techniques have been developed in the area of machine-learning and pattern recognition. In
this paper, we try to present the outline of the main stages through which a supervised
classification under the support vector machine algorithm is processed. We utilize a method
for applying SVM learning algorithm for data classification and regression purposes under a
cloud computing platform. We also point out that we have taken benefit of the various
advantages offered by the Google Earth Engine platform through which we have found that in
addition to guaranteed saving of time and effort, this platform has produced perfect results of
high precision. The latter can be shown in two precision indicators that we used and which
yielded very satisfactory results.
|