Applying Machine Learning Techniques to Transportation Mode Recognition Using Mobile Phone Sensor Data

This paper adopts different supervised learning methods from the field of machine learning to develop multiclass classifiers that identify the transportation mode, including driving a car, riding a bicycle, riding a bus, walking, and running. Methods that were considered include K-nearest neighbor, support vector machines (SVMs), and tree-based models that comprise a single decision tree, bagging, and random forest (RF) methods. For training and validating purposes, data were obtained from smartphone sensors, including accelerometer, gyroscope, and rotation vector sensors. K-fold cross-validation as well as out-of-bag error was used for model selection and validation purposes. Several features were created from which a subset was identified through the minimum redundancy maximum relevance method. Data obtained from the smartphone sensors were found to provide important information to distinguish between transportation modes. The performance of different methods was evaluated and compared. The RF and SVM methods were found to produce the best performance. Furthermore, an effort was made to develop a new additional feature that entails creating a combination of other features by adopting a simulated annealing algorithm and a random forest method.

I. Introduction

Distinguishing between different types of physical activities using sensor data has been a recent research topic that has received considerable attention [1], [2]. Transportation mode detection can be considered as an activity recognition task in which data from smartphone sensors carried by users are utilized to infer what transportation mode the individuals have used. Micro-electromechanical systems (MEMS), such as accelerometers and gyroscopes are embedded in most smartphone devices [3] from which the data can be obtained at high frequencies. Smartphones, nowadays, are equipped with powerful sensors such as GPS, accelerometer, gyroscope, light sensors, etc. Having such powerful sensors all embedded in a small device carried in everyday life activities has enabled researchers to investigate new research areas. The advantages of these smart devices include ubiquity, ability to send and receive data through various ways (e.g., Wi-Fi/cellular network/Bluetooth), and storing/processing data [4].