Abstract
Human Activity Recognition (HAR) using sensors offers a wide range of applications in healthcare, smart homes for the elderly, sports, and other fields. There have been numerous attempts in existing methods to distinguish diverse human actions using sensor data but all those studies are based on clean data patterns. As a result, investigation of detecting activity based on sensor values even if some data is lost in a random pattern. In this paper, the authors present a strategy for improving activity detection. A random pattern (Missing at random) for missing data since it is a realistic problem for sensor-based activities. First, recognize the activity using the clean dataset (HAR) and then generate different percentages of missing data in test and train datasets. At the time of classification of various activities, if a classifier is learned with missing data, it reinforces the model to regulate the missing data. This method proves that machine learning models plausibility by demonstrating that this kind of approach can predict the activity. In machine learning models, random forest and logistic regression are considered for activity classification. Using the mice algorithm in the training dataset authors observed that it can effectively boost recognition accuracy from 87% to 98%.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ranger, S.: What is the IoT? Everything you need to know about the Internet of Things right now (2020). https://www.zdnet.com/article/what-is-the-internet-of-things-everything-you-need-to-knoabout-the-iot-right-now/
Bhandari, P.: Missing Data—Types, Explanation, & Imputation (2021). https://www.scribbr.com/statistics/missing-data
Multiple Imputation by Chained Equations (MICE) Explained (2022). https://stats.stackexchange.com/questions/421545/multiple-imputation-by-chained-equations-mice-explained]
Kusuma, W.A., Minarno, A.E., Wibowo, M.S.: “Triaxial accelerometer-based human activity recognition using 1D convolution neural network. In: 2020 International Workshop on Big Data and Information Security (IWBIS), pp. 53–58 (2020). https://doi.org/10.1109/IWBIS50925.2020.9255581
Hossain, T., Goto, H., Inoue, S.: Improving activity recognition for missing data. In: Proceedings of the EAI International Conference on Mobile Computing, Applications and Services (MobiCASE) Student Workshop, Osaka, Japan, 28 February–2 March (2018)
Breiman, L.: Random forests. Mach. Learn. 45, 5–32 (2001)
Ogawa, N., Kaji, K., Kawaguchi, N.: Effects of number of subjects on activity recognition—findings from HASC2010corpus. In: Proceedings of the International Workshop on Frontiers in Activity Recognition using Pervasive Sensing (IWFAR2011), San Francisco, CA, USA, pp. 48–51 (2011)
Kawaguchi, N., et al.: HASC challenge: gathering large scale human activity corpus for the real-world activity understandings. In: Proceedings of the 2nd Augmented Human International Conference, Tokyo, Japan, pp. 27:1–27:5 (2011)
Casale, P., Pujol, O., Radeva, P.: Personalization and user verification in wearable systems using biometric walking patterns. Pers. Ubiquitous Comput 16, 563–580 (2012). https://doi.org/10.1007/s00779-011-0415-z
Cui, Z., Chen, W., He, Y., Chen, Y.: Optimal action extraction for random forests and boosted trees. In: Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sydney, Australia, pp. 179–188 (2015)
Dwivedi, A.D., Srivastava, G., Dhar, S., Singh, R.: A decentralized privacy-preserving healthcare blockchain for IoT. Sensors (Switzerland), 19(2), 1–17 (2019). https://doi.org/10.3390/s19020326
Kunta, V., Tuniki, C., Sairam, U.: Multi-functional blind stick for visually impaired people. In: 2020 5th International Conference on Communication and Electronics Systems (ICCES), pp. 895–899 (2020). https://doi.org/10.1109/ICCES48766.2020.9137870
Gao, Z., Cheng, W., Qiu, X., Meng, L.: A missing sensor data estimation algorithm based on temporal and spatial correlation. Int. J. Distrib. Sens. Netw 11, 435391 (2015)
Prabowo, O.M., Mutijarsa, K., Supangkat, S.H.: Missing data handling using machine learning for human activity recognition on mobile device. In: Proceedings of the International Conference on ICT for Smart Society, Surabaya, Indonesia (2016)
Pedro, J., Laencina, G., Luis, J., Gomez, S.: Pattern classification with missing data: a review. Springer, Germany (2009)
Chen, J.I.Z.: Smart security system for suspicious activity detection in volatile areas. J. Inf. Technol. 2(01), 64–72 (2020)
Bhambri, P., Bagga, S., Priya, D., Singh, H., Dhiman, H.K.: Suspicious human activity detection system. J. IoT Soc. Mobile Anal. Cloud 2(4), 216–221 (2020)
Adepu, Y., Boga, V.R., Sairam, U.: Interviewee performance analyzer using facial emotion recognition and speech fluency recognition. In: 2020 IEEE International Conference for Innovation in Technology (INOCON) (2020)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Guguloth, S., Telu, A., Sairam, U., Voruganti, S. (2023). Activity Recognition in Missing Data Scenario Using MICE Algorithm. In: Abraham, A., Hanne, T., Gandhi, N., Manghirmalani Mishra, P., Bajaj, A., Siarry, P. (eds) Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022). SoCPaR 2022. Lecture Notes in Networks and Systems, vol 648. Springer, Cham. https://doi.org/10.1007/978-3-031-27524-1_82
Download citation
DOI: https://doi.org/10.1007/978-3-031-27524-1_82
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-27523-4
Online ISBN: 978-3-031-27524-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)