Human activity recognition based on feature selection in smart home using back-propagation algorithm

• Published in: ISA transactions Vol. 53; no. 5; pp. 1629 - 1638
• Main Authors: Fang, Hongqing, He, Lei, Si, Hao, Liu, Peng, Xie, Xiaolei
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.09.2014
• Subjects: Accuracy; Actigraphy - methods; Algorithms; Artificial Intelligence; Bayes Theorem; Classifiers; Feature recognition; Feature selection; Human activity recognition; Human motion; Humans; Markov models; Monitoring, Ambulatory - methods; Motor Activity - physiology; Neural networks; Pattern Recognition, Automated - methods; Pervasive computing; Recognition; Reproducibility of Results; Sensitivity and Specificity; Sensors and networks; Smart home; Telemedicine - methods; Pervasive computing; Feature selection; Human activity recognition; Sensors and networks; Smart home
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2014.06.008

In this paper, Back-propagation(BP) algorithm has been used to train the feed forward neural network for human activity recognition in smart home environments, and inter-class distance method for feature selection of observed motion sensor events is discussed and tested. And then, the human activity recognition performances of neural network using BP algorithm have been evaluated and compared with other probabilistic algorithms: Naïve Bayes(NB) classifier and Hidden Markov Model(HMM). The results show that different feature datasets yield different activity recognition accuracy. The selection of unsuitable feature datasets increases the computational complexity and degrades the activity recognition accuracy. Furthermore, neural network using BP algorithm has relatively better human activity recognition performances than NB classifier and HMM. •Used BP algorithm to train feed forward neural network for activity recognition.•The suitable feature set must be selected in advance.•Adopted Inter-class distance method for feature selection of motion sensor events.•Compared the performance measures of BPNN with those of NB and HMM.•The performance measures of BPNN are better than those of NB and HMM.