Human Activity Recognition Using Smartphone Dataset

Abstract

This project depicts recognition of Human activities using data generated from user’s Smart phone. We have used data available at University of California Machine Learning repository to recognize six human activities. These activities are Standing, Sitting, Laying, Walking, Walking upstairs and Walking downstairs. Data is collected from embedded accelerometer, gyroscope and other sensors of Samsung Galaxy S II Smart phone. Data is randomly divided into 7:3 ratios to form training and testing data set respectively. Dimensionality reduction is done using Principal Component Analysis technique. Activity classification is done using Machine Learning models namely Random Forest, Support Vector Machine, Artificial Neural Network and K-Nearest Neighbour. We have compared accuracy and performance of these models using confusion matrix and random simulation.