Optimal model selection for posture recognition in home-based healthcare

This paper investigates optimal model selection for posture recognition. Accuracy and computational time are related to the trained model in a supervised classification. An optimal model selection is important for a reliable activity monitoring system. Conventional guidance on model training uses large instances of randomly selected data in order to characterize the classes. A new approach to the training of a multiclass support vector machine (SVM) model suited to limited training sets such as used in posture recognition is provided. This approach picks a small training set from misclassified data to improve an initial model in an iterative and incremental fashion. In addition, a two step grid-search algorithm is used for the parameters setting. The best parameters were chosen according to the testing accuracy rather than conventional validating accuracy. This new approach for model selection was evaluated against conventional approaches in an activity classification study. Nine everyday postures were classified from a belt-worn smart phone’s accelerometer data. The classification derived from the small training set and the conventional randomly selected training set differed in two aspects: classification performance to new data (85.1% Pick-out small training set vs. 70.3% conventional large training set) and computational efficiency (improved 28%).

[1]  B. G. Celler,et al.  Classification of basic daily movements using a triaxial accelerometer , 2004, Medical and Biological Engineering and Computing.

[2]  M. Kolehmainen,et al.  Cow behaviour pattern recognition using a three-dimensional accelerometer and support vector machines , 2009 .

[3]  Shigeo Abe,et al.  Fast Training of Support Vector Machines by Extracting Boundary Data , 2001, ICANN.

[4]  Gerhard Tröster,et al.  SMASH: a distributed sensing and processing garment for the classification of upper body postures , 2008, BODYNETS.

[5]  Pedro M. Domingos,et al.  On the Optimality of the Simple Bayesian Classifier under Zero-One Loss , 1997, Machine Learning.

[6]  Alex Pentland,et al.  Wearable feedback systems for rehabilitation , 2005, Journal of NeuroEngineering and Rehabilitation.

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

[8]  Ian H. Witten,et al.  Data mining - practical machine learning tools and techniques, Second Edition , 2005, The Morgan Kaufmann series in data management systems.

[9]  R. Kass,et al.  Multiple neural spike train data analysis: state-of-the-art and future challenges , 2004, Nature Neuroscience.

[10]  J. B. J. Bussmann,et al.  Measuring daily behavior using ambulatory accelerometry: The Activity Monitor , 2001, Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc.

[11]  Eibe Frank,et al.  Combining Naive Bayes and Decision Tables , 2008, FLAIRS.

[12]  Ingo Mierswa,et al.  YALE: rapid prototyping for complex data mining tasks , 2006, KDD '06.

[13]  R. Fletcher.  Practical Methods of Optimization , 1988 .

[14]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[15]  Pat Morin,et al.  Output-Sensitive Algorithms for Computing Nearest-Neighbour Decision Boundaries , 2005, Discret. Comput. Geom..

[16]  Rita Cucchiara,et al.  Probabilistic posture classification for Human-behavior analysis , 2005, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[17]  Friedrich Foerster,et al.  Detection of posture and motion by accelerometry : a validation study in ambulatory monitoring , 1999 .

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

[19]  Gérard Dreyfus,et al.  Single-layer learning revisited: a stepwise procedure for building and training a neural network , 1989, NATO Neurocomputing.

[20]  S. Halgamuge,et al.  Reducing the Number of Training Samples for Fast Support Vector Machine Classification , 2004 .

[21]  V. Vapnik.  Pattern recognition using generalized portrait method , 1963 .

[22]  HyungDal Park,et al.  A study on the activity classification using a triaxial accelerometer , 2003, Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439).

[23]  Rita Cucchiara,et al.  A machine learning approach for human posture detection in domotics applications , 2003, 12th International Conference on Image Analysis and Processing, 2003.Proceedings..

[24]  G. Tröster,et al.  Sensor for Measuring Strain in Textile , 2008, Sensors.

[25]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[26]  Chia-Feng Juang,et al.  Human Body Posture Classification by a Neural Fuzzy Network and Home Care System Application , 2007, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[27]  T. M. Williams,et al.  Practical Methods of Optimization. Vol. 1: Unconstrained Optimization , 1980 .

[28]  Shigeo Abe.  Support Vector Machines for Pattern Classification , 2010, Advances in Pattern Recognition.

[29]  François Brémond,et al.  Applying 3D human model in a posture recognition system , 2006, Pattern Recognit. Lett..

[30]  Eliathamby Ambikairajah,et al.  Classification of a known sequence of motions and postures from accelerometry data using adapted Gaussian mixture models. , 2006, Physiological measurement.

[31]  O Karhu,et al.  Correcting working postures in industry: A practical method for analysis. , 1977, Applied ergonomics.

[32]  Michael L. Littman,et al.  Activity Recognition from Accelerometer Data , 2005, AAAI.

[33]  Huiru Zheng,et al.  A theoretic algorithm for fall and motionless detection , 2009, 2009 3rd International Conference on Pervasive Computing Technologies for Healthcare.