Improving Accelerometer-Based Activity Recognition by Using Ensemble of Classifiers
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2016, Vol 7, Issue 5
Abstract
In line with the increasing use of sensors and health application, there are huge efforts on processing of collected data to extract valuable information such as accelerometer data. This study will propose activity recognition model aim to detect the activities by employing ensemble of classifiers techniques using the Wireless Sensor Data Mining (WISDM). The model will recognize six activities namely walking, jogging, upstairs, downstairs, sitting, and standing. Many experiments are conducted to determine the best classifier combination for activity recognition. An improvement is observed in the performance when the classifiers are combined than when used individually. An ensemble model is built using AdaBoost in combination with decision tree algorithm C4.5. The model effectively enhances the performance with an accuracy level of 94.04 %.
Authors and Affiliations
Tahani Daghistani, Riyad Alshammari
Keywords
Related Articles
- EP ID EP96119
- DOI 10.14569/IJACSA.2016.070520
- Views 125
- Downloads 0
A New Motion Planning Framework based on the Quantized LQR Method for Autonomous Robots
This study addresses an argument on the disconnection between the computational side of the robot navigation problem with the control problem including concerns on stability. We aim to constitute a framework that include...
Design of High Precision Temperature Measurement System based on Labview
Using the LabVIEW software platform, a high precision temperature measuring device is designed based on the principle of the thermocouple. The system uses the STM32 MCU as the main control chip, using AD7076 analog digit...
Policy-Based Automation of Dynamique and Multipoint Virtual Private Network Simulation on OPNET Modeler
The simulation of large-scale networks is a challenging task especially if the network to simulate is the Dynamic Multipoint Virtual Private Network, it requires expert knowledge to properly configure its component techn...
Linear Prediction Model for Effort in Programming based on User Acceptance and Revised use Case Point Method
As long as most of the processes of verification and validation of software to grant acceptance by the customer/user, are subjective type, it is aimed to design a standard mathematical model with empirical to perform an...
Formal Method to Derive Interoperability Requirements and Guarantees
Interoperability among telecommunications systems, possibly by different vendors, is essential for both the development of many telecommunications networks, and today's civilization development. Interoperability testing...