Recently, many wearable devices have been developed as IoT technology grows. Among them, smart watch is the friendliest wearable device in daily lives. Many companies are trying to improve the device or system to provide personal service as user’s behavior. This paper proposes an user behavior classification system using smart watch and machine learning algorithm to provide personal service with wearable devices. Sensing data from accelerometer of smart watch is collected, and then classification is implemented by Two-Class Support Vector Machine, Multi-Class Logistic Regression, Multi-Class Decision Forest. We show that the prediction accuracy is more than 90%.
Paper
Full text
User behavior classification based on smart watch and machine learning algorithm
Semantic Scholar · Computer Science · 2017
Abstract
Recently, many wearable devices have been developed as IoT technology grows. Among them, smart watch is the friendliest wearable device in daily lives. Many companies are trying to improve the device or system to provide personal service as user’s behavior. This paper proposes an user behavior classification system using smart watch and machine learning algorithm to provide personal service with wearable devices. Sensing data from accelerometer of smart watch is collected, and then classification is implemented by Two-Class Support Vector Machine, Multi-Class Logistic Regression, Multi-Class Decision Forest. We show that the prediction accuracy is more than 90%.