An intelligent framework to determine a mobile device context utilizing in-built sensors

Rapid proliferation of numerous sensing technologies into mobile devices making them smarter than ever before. The smartphone is now better able to sense and keep record of various activities such as fitness level, mood detection, individual's drinking state, driving behavior, personal interests etc. Having such varied source of information can my smartphone really act as a recommender or as an assistant, for example, a mobile device can alert for a nearby gas station if the vehicle has driven for long. Likewise, a construction alert on the way to workplace, if any, can be displayed on the user's mobile device. Thus, to really make the smartphone more of an assistant rather than user-operated device, knowing the context of a user and mobile device becomes critically important. The research work presented in this study develops an intelligent framework able to sense the mobile device context; that is the placement of a device carried by a user (e.g., shirt pocket, trouser pocket, handheld etc.). The framework extracts the features out of accelerometer and gyroscope measurements and further derive the mobile device context using an offline trained machine learning module. For mobile device context classification we considered Support Vector Machines (SVMs) due to its improved generalization capability. The framework when tested over random participant was able to classify with an accuracy of above 92%.

[1]  R. Brereton,et al.  Support vector machines for classification and regression. , 2010, The Analyst.

[2]  S.Y. Lee,et al.  Accelerometer's position free human activity recognition using a hierarchical recognition model , 2010, The 12th IEEE International Conference on e-Health Networking, Applications and Services.

[3]  Diogo R. Ferreira,et al.  Preprocessing techniques for context recognition from accelerometer data , 2010, Personal and Ubiquitous Computing.

[4]  Ana M. Bernardos,et al.  Activity logging using lightweight classification techniques in mobile devices , 2012, Personal and Ubiquitous Computing.

[5]  Lijuan Cao,et al.  A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine , 2003, Neurocomputing.

[6]  Emiliano Miluzzo,et al.  Pocket, Bag, Hand, etc. - Automatically Detecting Phone Context through Discovery , 2010 .

[7]  Doruk Coskun,et al.  Phone position/placement detection using accelerometer: Impact on activity recognition , 2015, 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP).

[8]  Y. Kawahara,et al.  Recognizing User Context Using Mobile Handsets with Acceleration Sensors , 2007, 2007 IEEE International Conference on Portable Information Devices.

[9]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.