Comparative Analysis of Classification Algorithms using UCI Data Set based on Smartphone Accelerometer and Gyroscope

Mobile phone based activity recognition uses dataobtained from embedded sensors to infer user’sphysical activities. Therefore, many mobile phoneshave been equipped with sensors to enable thedelivery of advanced features to the users.Accelerometer and gyroscope are the sensors thatembedded to several types of mobiles devices. In thispaper, we apply 17 classifier algorithms to select thebest performance ones using UCI data sets. Thesedataset are labeled twelve human activities. To testthe performance accuracy of these algorithms, the10-fold cross validation is done using Weka 3.6.11data mining tool. The overall accuracy rates forclassifiers are exceeded 85% and nearly 96% whichare encouraged results. Thus, we select theappropriate classifier algorithms based on theseaccuracy results to be used for online human activity recognition