A Context-Aware Motion Mode Recognition System Using Embedded Inertial Sensors in Portable Smart Devices Omar Sheishaa, Mohamed Tamazin, and Iman Morsi Abstract The expeditious market transformation to smart portable devices has created an opportunity to support activity recognition using the embedded sensors of these devices. Over the last decade, many activity recognition approaches have been proposed for various activities in different settings. The motion mode recog- nition or transition in modes of the device is needed in many technological domains. This approach detects a variety of motion modes for a human using a portable device. The approach includes many aspects: usability, mounting and data acquisi- tion, sensors used, signal processing, methods employed, features extracted, and classification techniques. This chapter sums up with a comparison of the perfor- mance of several motion mode recognition techniques. In this research, multiple behaviors were distinguished using embedded inertial sensors in portable smart devices. In our experiments, we selected four types of human activity, which are walking, standing, sitting, and running. A combination of one of the embedded mobile sensors and machine learning techniques have been proposed in order to do this kind of classification. The proposed system relies on accelerometer data to classify user activities. The results show that using SVM classifier showed better accuracy for detection compared to the outcomes of the other classifiers like KNN and ensemble classifiers. For future work, classification of other human activities like cycling, driving, and swimming will be investigated. Keywords Motion mode recognition · Machine learning · Smartphone accelerometer data O. Sheishaa (*) · M. Tamazin · I. Morsi Electronics and Communications Engineering Department, Arab Academy for Science, Technology and Maritime Transport, Alexandria, Egypt © Springer Nature Switzerland AG 2020 M. H. Farouk, M. A. Hassanein (eds.), Recent Advances in Engineering Mathematics and Physics, https://doi.org/10.1007/978-3-030-39847-7_23 275