Accelerometer-based human activity classification using Water Wave Optimization approach

This paper introduces an approach of classifying accelerometer data for simple human activities using Support Vector Machine (SVM). Classifier results are being optimized by (WWO) Water Wave Optimization Algorithm. Human activity classification has been a hot research point and has been an aid for smart human activity recognition-based systems and for analysis purposes. Numerous classification methods are introduced by researchers including decision trees, bagging of trees, boosting of trees and random forests [1] as well as the SVM that is employed in this work. Accelerometers are very effective source of data for human activity recognition purposes and have been used in many research efforts as well as this paper. This work shows the process of optimizing SVM parameters in order to get better classification results for a set of human activity accelerometer-based data. The applied optimization algorithm is (WWO) [2], a meta-heuristic evolutionary algorithm. 15-Fold cross validation is accomplished with different data preparation using a leave-one-subject-out approach. Classification accuracy on the applied dataset ranges from 81% to 97% for a first run, and approached 100% accuracy for a second run. validation being applied on 15 different folds on 15 separate dataset files each represents a different participant. We employ four different data preparation configurations.

[1]  Kenji Mase,et al.  Activity and Location Recognition Using Wearable Sensors , 2002, IEEE Pervasive Comput..

[2]  Johannes Peltola,et al.  Activity classification using realistic data from wearable sensors , 2006, IEEE Transactions on Information Technology in Biomedicine.

[3]  J.K. Aggarwal,et al.  Human activity analysis , 2011, ACM Comput. Surv..

[4]  Albrecht Schmidt,et al.  Multi-sensor Activity Context Detection for Wearable Computing , 2003, EUSAI.

[5]  Sheng Gao,et al.  Classifier Optimization for Multimedia Semantic Concept Detection , 2006, 2006 IEEE International Conference on Multimedia and Expo.

[6]  Petia Radeva,et al.  Human Activity Recognition from Accelerometer Data Using a Wearable Device , 2011, IbPRIA.

[7]  Tatsuo Nakajima,et al.  Feature Selection and Activity Recognition from Wearable Sensors , 2006, UCS.

[8]  Amir Hossein Gandomi,et al.  Bat algorithm for constrained optimization tasks , 2012, Neural Computing and Applications.

[9]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[10]  P. J. García Nieto,et al.  A hybrid PSO optimized SVM-based model for predicting a successful growth cycle of the Spirulina platensis from raceway experiments data , 2016, J. Comput. Appl. Math..

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

[12]  José Sergio Ruiz Castilla,et al.  PSO-Based Method for SVM Classification on Skewed Data-Sets , 2015, ICIC.

[13]  Patrick Siarry,et al.  A survey on optimization metaheuristics , 2013, Inf. Sci..

[14]  Douglas M. Hawkins,et al.  The Problem of Overfitting , 2004, J. Chem. Inf. Model..

[15]  Yujun Zheng.  Water wave optimization: A new nature-inspired metaheuristic , 2015, Comput. Oper. Res..

[16]  Miguel A. Labrador,et al.  A Survey on Human Activity Recognition using Wearable Sensors , 2013, IEEE Communications Surveys & Tutorials.

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

[18]  Petia Radeva,et al.  Personalization and user verification in wearable systems using biometric walking patterns , 2011, Personal and Ubiquitous Computing.

[19]  Erik Valdemar Cuevas Jiménez,et al.  A swarm optimization algorithm inspired in the behavior of the social-spider , 2013, Expert Syst. Appl..

[20]  Dong Ho Song,et al.  Optimizing Weighted Kernel Function for Support Vector Machine by Genetic Algorithm , 2006, MICAI.

[21]  Graeme Hirst,et al.  Identifying Sexual Predators by SVM Classification with Lexical and Behavioral Features , 2012, CLEF.

[22]  Beatrice Lazzerini,et al.  A new multi-objective evolutionary algorithm based on convex hull for binary classifier optimization , 2007, 2007 IEEE Congress on Evolutionary Computation.

[23]  Jaime G. Carbonell,et al.  On the parameter optimization of Support Vector Machines for binary classification. , 2012, Journal of integrative bioinformatics.

[24]  Xi Long,et al.  Single-accelerometer-based daily physical activity classification , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[25]  Michael L. Raymer,et al.  GA-facilitated KNN classifier optimization with varying similarity measures , 2005, 2005 IEEE Congress on Evolutionary Computation.

[26]  Amir Hossein Gandomi,et al.  Cuckoo search algorithm: a metaheuristic approach to solve structural optimization problems , 2011, Engineering with Computers.

[27]  Zhongyi Hu,et al.  A PSO and pattern search based memetic algorithm for SVMs parameters optimization , 2013, Neurocomputing.

[28]  Weizhong Zhao,et al.  Detecting Lung Diseases from Exhaled Aerosols: Non-Invasive Lung Diagnosis Using Fractal Analysis and SVM Classification , 2015, PloS one.

[29]  A. Mucherino,et al.  Monkey search: a novel metaheuristic search for global optimization , 2007 .

[30]  Abdulhamit Subasi,et al.  Classification of EMG signals using PSO optimized SVM for diagnosis of neuromuscular disorders , 2013, Comput. Biol. Medicine.

[31]  Pedro Ribeiro,et al.  Human Activity Recognition from Video: modeling, feature selection and classification architecture , 2005 .