International Journal of Knowledge Based Computer System

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Efficient Classification Techniques of Human Activities from Smartphone Sensor Data using Machine Learning Algorithms


Affiliations
1 Department of Computer Science & Engineering, Islamic University, Kushtia, Bangladesh
     

   Subscribe/Renew Journal


Increasing use of accelerometer and protractor sensors in recent years has created a field of study for the definition of human activities. This issue is tried to be solved by using machine learning methods. For this, it is solved by extracting different properties from the obtained signals, obtaining the characteristics specific to the activity and classifying these properties. In this study, the time and frequency domain properties of 4 different human activities were extracted, then a pre-treatment step was applied in accordance with the obtained feature set, and then the size was reduced with PCA and Fisher ‘LDA methods. The k-NN classifier and perceptron classifiers were designed for the obtained feature set and the classification process was performed. In this study, the classification success of these methods using different parameters has been examined and the results are shown.

Keywords

Feature Extraction, Fisher’s LDA, Gradient Descent Method, Human Activity Identification, Kessler’s Reconstruction, k-NN Classifier, PCA
Subscription Login to verify subscription
User
Notifications
Font Size


  • J. Morales, and D. Akopian, “Physical activity recognition by smartphones, a survey,” Biocybernetics and Biomedical Engineering, vol. 37, no. 3, pp. 388-400, 2017, doi: 10.1016/j.bbe.2017.04.004.

  • J. Qi, P. Yang, A. Waraich, Z. Deng, Y. Zhao, and Y. Yang, “Examining sensor-based physical activity recognition and monitoring for healthcare using internet of things: A systematic review,” Journal of Biomedical Informatics, vol. 87, pp. 138-153, 2018, doi: 10.1016/j.jbi.2018.09.002.

  • R.-A. Voicu, C. Dobre, L. Bajenaru, and R.-I. Ciobanu, “Human physical activity recognition using smartphone sensors,” Sensors, vol. 19, no. 3, p. 458, 2019, doi: 10.3390/s19030458.

  • A. Bayat, M. Pomplun, and D. A. Tran, “A study on human activity recognition using accelerometer data from smartphones,” Procedia Computer Science, vol. 34, pp. 450-457, 2014, doi: 10.1016/j.procs.2014.07.009.

  • P. Casale, O. Pujol, and P. Radeva, “Human activity recognition from accelerometer data using a wearable device,” Pattern Recognit. Image Anal., 6669, 2011, pp. 289-296, doi: 10.1007/978-3-642-21257-4_36.

  • S. Mysore, “Human activity recognition from sensor data,” 2018.

  • J. Lester, T. Choudhury, and G. Borriello, “A practical approach to recognizing physical activities,” Pervasive, 2006.

  • W. Xingfeng, and K. Heecheol, “Detecting user activities with the accelerometer on android smartphones,” Journal of Mutimedia Information System, vol. 2, no. 2, pp. 207-214, 2015.

  • D. M. Pober, J. Staudenmayer, C. Raphael, and P. S. Freedson, “Development of novel techniques to classify physical activity mode using accelerometers,” Med Sci Sports Exerc., vol. 38, no. 9, pp. 1626-1634, 2006.

  • J.-H. Cho, J. T. Kim, and T.-S. Kim, “Smart phone-based human activity classification and energy expenditure generation in building environments,” Proc. 7th Int. Symp. Sustain. Healthy Buildings, 2012.

  • N. Ravi, N. Dandekar, P. Mysore, and M. L. Littman, “Activity recognition from accelerometer data,” AAAI, 3, pp. 1541-1546, 2005.

  • A. Mannini, and A. M. Sabatini, “Machine learning methods for classifying human physical activity from on-body accelerometers,” Sensors, vol. 10, no. 2, pp. 1154-1175, 2010, doi: 10.3390/s100201154.

  • H. Chen, S. Mahfuz, and F. Zulkernine, “Smart phone based human activity recognition,” 2019 IEEE Int. Conf. Bioinformatics and Biomedicine (BIBM), 2019, pp. 2525-2532, doi: 10.1109/BIBM47256.2019.8983009.

  • https://bmi.hmu.gr/the-mobifall-and-mobiact-datasets-2/

  • D. Micucci, M. Mobilio, and P. Napoletano, “UniMiB SHAR: A new dataset for human activity recognition using acceleration data from smartphones,” Applied Sciences, vol. 7, no. 10, p. 1101, 2017, doi: 10.3390/app7101101.

  • G. Vavoulas, C. Chatzaki, T. Malliotakis, M. Pediaditis, and M. Tsiknakis, “The MobiAct dataset: Recognition of activities of daily living using smartphones,” 2nd Int. Conf. Inf. and Commun. Technologies for Ageing Well and e-Health, 2016, pp. 143-151, doi: 10.5220/0005792401430151.

  • S. Fan, Y. Jia, and C. Jia, “A feature selection and classification method for activity recognition based on an inertial sensing unit,” Information, vol. 10, no. 10, p. 290, 2019, doi: 10.3390/info10100290.

  • S. J. Preece, J. Y. Goulermas, L. P. J. Kenney, D. Howard, K. Meijer, and R. H. Crompton, “Activity identification using body-mounted sensors - A review of classification techniques,” Physiological Measurement, vol. 30, no. 4, R1-33, 2009, doi: 10.1088/0967-3334/30/4/R01.

  • S. Rosati, G. Balestra, and M. Knaflitz, “Comparison of different sets of features for human activity recognition by wearable sensors,” Sensors, vol. 18, no. 12, p. 4189, 2018, doi: 10.3390/s18124189.

  • https://actigraphcorp.force.com/support/s/article/What-is-the-feature-extraction-tool-and-how-does-it-work

  • J. Zhu, R. San-Segundo, and J. M. Pardo, “Feature extraction for robust physical activity recognition,” Human-Centric Computing and Information Sciences, vol. 7, 2017, Art. no. 16, doi: 10.1186/s13673-017-0097-2.

  • B. Pappachan, W. Caesarendra, T. Tjahjowidodo, and T. Wijaya, “Frequency domain analysis of sensor data for event classification in real-time robot assisted deburring,” Sensors, vol. 17, no. 6, p. 1247, 2017, doi: 10.3390/s17061247.

  • Md. Aktaruzzaman, N. Scarabottolo, and R. Sassi, “Parametric estimation of sample entropy for physical activity recognition,” 2015 37th Annu. Int. Conf. IEEE Eng. Medicine and Biol. Soc. (EMBC), Milan, Italy, Aug. 25-29, 2015, doi: 10.1109/EMBC.2015.7318401.

  • C. Reining, F. M. Rueda, M. ten Hompel, G. A. Fink, and F. Niemann, “Human activity recognition for production and logistics - A systematic literature review,” Information, vol. 10, no. 8, p. 245, 2019, doi: 10.3390/info10080245.

  • I. M. Pires, N. Garcia, N. Pombo, F. Flórez-Revuelta, S. Spinsante, M. C. C. Teixeira, and E. Zdravevski, “Pattern recognition techniques for the identification of activities of daily living using mobile device accelerometer,” 2018, doi: 10.7287/peerj.preprints.27225.

  • Y.-L. Hsu, S.-C. Yang, H.-C. Chang, and H.-C. Lai, “Human daily and sport activity recognition using a wearable inertial sensor network,” IEEE Access, pp. 1-1, 2018, doi: 10.1109/ACCESS.2018.2839766.

  • B. Prapochanang, T. Leauhatong, and N. Nattaitanakul, “A novel dimension reduction technique for physical activity recognition from a waist-mounted accelerometer,” pp. 532-536, 2016, doi: 10.12792/iciae2016.097.

  • I. E. Moudden, H. Jouhari, S. E. Bernoussi, and M. Ouzir, “Learned model for human activity recognition based on dimensionality reduction,” 2018.

  • I. E. Moudden, Benyacoub, and S. E. Bernoussi, “Modeling human activity recognition by dimensionality reduction approach,” 2016.

  • M. Abidine, B. Fergani, and I. Menhour, “Activity recognition from smartphones using hybrid classifier PCA-SVM-HMM,” pp. 1-5, 2019, doi: 10.1109/WINCOM47513.2019.8942492.

  • M. Abidine, and B. Fergani, “Evaluating a new classification method using PCA to human activity recognition,” Int. Conf. Comput. Medical Appl. (ICCMA’ 2013), 2013, doi: 10.1109/ICCMA.2013.6506158.

  • K. Walse, R. V. Dharaskar, and V. M. Thakare, “PCA based optimal ANN classifiers for human activity recognition using mobile sensors data,” 2016, doi: 10.1007/978-3-319-30933-0_43.

  • Y. Saez, A. Baldominos, and P. Isasi, “A comparison study of classifier algorithms for cross-person physical activity recognition,” Sensors, vol. 17, no. 1, p. 66, 2017, doi: 10.3390/s17010066.

  • X. Xi, M. Tang, S. M. Miran, and Z. Luo, “Evaluation of feature extraction and recognition for activity monitoring and fall detection based on wearable sEMG sensors,” Sensors, vol. 17, no. 6, p. 1229, 2017, doi: 10.3390/s17061229.

  • A.-M. Mandong, and U. Munir, “Smartphone based activity recognition using K-Nearest Neighbor algorithm,” 2018.

  • S. Shalev-Shwartz, “Perceptron algorithm,” 2008, doi: 10.1007/978-0-387-30162-4_287.


Abstract Views: 163

PDF Views: 0




  • Efficient Classification Techniques of Human Activities from Smartphone Sensor Data using Machine Learning Algorithms

Abstract Views: 163  |  PDF Views: 0

Authors

Diponkor Bala
Department of Computer Science & Engineering, Islamic University, Kushtia, Bangladesh
G. M. Waliullah
Department of Computer Science & Engineering, Islamic University, Kushtia, Bangladesh

Abstract


Increasing use of accelerometer and protractor sensors in recent years has created a field of study for the definition of human activities. This issue is tried to be solved by using machine learning methods. For this, it is solved by extracting different properties from the obtained signals, obtaining the characteristics specific to the activity and classifying these properties. In this study, the time and frequency domain properties of 4 different human activities were extracted, then a pre-treatment step was applied in accordance with the obtained feature set, and then the size was reduced with PCA and Fisher ‘LDA methods. The k-NN classifier and perceptron classifiers were designed for the obtained feature set and the classification process was performed. In this study, the classification success of these methods using different parameters has been examined and the results are shown.

Keywords


Feature Extraction, Fisher’s LDA, Gradient Descent Method, Human Activity Identification, Kessler’s Reconstruction, k-NN Classifier, PCA

References