<![CDATA[Human Activity Recognition is a technology that introduces human body movement using an accelerometer, gyroscope, global positioning system, and camera. The early emergence of the support vector machine method was used to classify 2 classes, so development was needed to overcome multiclass problems and a large number of large-scale datasets resulted in suboptimal performance. The purpose of this paper is to apply the ensemble Support Vector Machine method in classifying the movement of walking, running, and climbing stairs based on accelerometer and gyroscope sensors on smartphones. And see the performance of the Ensemble Support Vector Machine method when using linear kernels and RBF. The results of the Support Vector Machine linear kernel accuracy of 79.66% and an increase of 88.01% after using the ensemble. While the accuracy for the Support Vector Machine kernel RBF is 79.51 and an increase of 88.04% after using the ensemble.]]>
Copyrights © 2022
