A Mobile Application to Detect Abnormal Patterns of Activity

In this paper we introduce an unsupervised online clustering algorithm to detect abnormal activities using mobile devices. This algorithm constantly monitors a user’s daily routine and builds his/her personal behavior model through online clustering. When the system observes activities that do not belong to any known normal activities, it immediately generates alert signals so that incidents can be handled in time. In the proposed algorithm, activities are characterized by users’ postures, movements, and their indoor location. Experimental results show that the behavior models are indeed user-specific. Our current system achieves 90% precision and 40% recall for anomalous activity detection.

[1]  Tapio Seppänen,et al.  Recognizing human motion with multiple acceleration sensors , 2001, 2001 IEEE International Conference on Systems, Man and Cybernetics. e-Systems and e-Man for Cybernetics in Cyberspace (Cat.No.01CH37236).

[2]  A. Nguyen,et al.  Unsupervised Clustering of Free-Living Human Activities using Ambulatory Accelerometry , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.

[4]  C. Randell,et al.  Context awareness by analysing accelerometer data , 2000, Digest of Papers. Fourth International Symposium on Wearable Computers.

[5]  Andreas Krause,et al.  Unsupervised, dynamic identification of physiological and activity context in wearable computing , 2003, Seventh IEEE International Symposium on Wearable Computers, 2003. Proceedings..

[6]  Thomas Kirste,et al.  Towards Recognizing Abstract Activities: An Unsupervised Approach , 2008, BMI.

[7]  K. Aminian,et al.  Physical activity monitoring based on accelerometry: validation and comparison with video observation , 1999, Medical & Biological Engineering & Computing.

[8]  Norbert Gyorbíró,et al.  Activity recognition system for mobile phones using the MotionBand device , 2008, MOBILWARE.

[9]  Philipp Bolliger,et al.  Redpin - adaptive, zero-configuration indoor localization through user collaboration , 2008, MELT '08.

[10]  Uwe Hansmann,et al.  Pervasive Computing , 2003 .

[11]  Ying Zhang,et al.  Proceedings of the first ACM international workshop on Mobile entity localization and tracking in GPS-less environments , 2008, MobiCom 2008.