17
2
Supervised and Semi-Supervised Identification
of Users and Activities from Wearable
Device Brain Signal Recordings
Glavin Wiechert, Matt Triff, Zhixing Liu, Zhicheng Yin, Shuai Zhao,
Ziyun Zhong, Runxing Zhou, and Pawan Lingras
Saint Mary’s University Halifax
Nova Scotia, Canada
CONTENTS
2.1 Introduction .......................................................................................................................... 18
2.2 Review of Wearables ........................................................................................................... 19
2.2.1 Wristbands and Watches ........................................................................................ 19
2.2.2 Armbands ................................................................................................................. 19
2.2.3 Headbands, Headsets, and Smartglasses ............................................................. 19
2.2.4 e-Textiles: Smart Clothing, Smart Textiles, Smart Fabrics ................................. 20
2.3 Review of Supervised Learning ........................................................................................ 20
2.3.1 Decision Tree ............................................................................................................ 20
2.3.2 Random Forest ......................................................................................................... 21
2.3.3 Support Vector Machine ......................................................................................... 21
2.3.4 Neural Network ....................................................................................................... 21
2.4 Semi-Supervised Learning with Genetic Algorithms and Rough Set Theory ...........22
2.4.1 K-means Clustering .................................................................................................22
2.4.2 Adaptation of Rough Set Theory for Clustering .................................................22
2.4.3 Genetic Algorithms ................................................................................................. 23
2.4.4 Genetic Algorithms for Rough K-Medoid Clustering ........................................ 24
2.5 Study Data ............................................................................................................................. 26
2.6 Knowledge Representation ................................................................................................ 27
2.7 Experimental Design ........................................................................................................... 29
2.7.1 Data ............................................................................................................................ 29
2.7.2 Classification .............................................................................................................30
2.7.3 Semi-Supervised Evolutionary Learning .............................................................30
2.8 Classification Results ........................................................................................................... 31
2.9 Semi-Supervised Evolutionary Learning Results ........................................................... 33
2.10 Conclusion ............................................................................................................................ 38
References ....................................................................................................................................... 39