Recognition of Activities in Resource Constrained Environments; Reducing the Computational Complexity
In our current work we propose a strategy to reduce the vast amounts of data produced within smart environments for sensor-based activity recognition through usage of the nearest neighbor (NN) approach. This approach has a number of disadvantages when deployed in resource constrained environments due to its high storage requirements and computational complexity. These requirements are closely related to the size of the data used as input to NN. A wide range of prototype generation (PG) algorithms, which are designed for use with the NN approach, have been proposed in the literature to reduce the size of the data set. In this work, we investigate the use of PG algorithms and their effect on binary sensor-based activity recognition when using a NN approach. To identify the most suitable PG algorithm four datasets were used consisting of binary sensor data and their associated class activities. The results obtained demonstrated the potential of three PG algorithms for sensor-based activity recognition that reduced the computational complexity by up to 95 % with an overall accuracy higher than 90 %.
KeywordsActivity recognition Resource constrained environments Nearest Neighbor (NN) Prototype generation (PG) Computational complexity
This contribution has been supported by research projects: TIN2015-66524-P and UJAEN/2014/06/14. Invest Northern Ireland is acknowledged for partially supporting this project under the Competence Centre Program Grant RD0513853 - Connected Health Innovation Centre.
- 2.Maksimovic, M., Vujovic, V., Davidovic, N., Milosevic, V., Raspberry, P.B.: Pi as internet of things hardware: performances and constraints. Des. Issues 3(8), 8 (2014)Google Scholar
- 3.Doukas, C.: Building Internet of Things with the ARDUINO. CreateSpace Independent Publishing Platform, USA (2012)Google Scholar
- 5.Chui, M., Loffler, M., Roberts, R.: The Internet of Things (Process Optimization). McKinsey Quarterly, Germany (2010)Google Scholar
- 8.San Martin, L.A., Pelaez, V.M., Gonzalez, R., Campos, A., Lobato, V.: Environmental user-preference learning for smart homes: an autonomous approach. J. Ambient Intell. Smart Environ. 2(3), 327–342 (2010)Google Scholar
- 13.Al-Faiz, M.Z., Ali, A.A., Miry, A.H.: A k-nearest neighbor based algorithm for human arm movements recognition using EMG signals. In: EPC-IQ 2010, pp. 159–167 (2010)Google Scholar
- 15.Moayeri Pour, G., Troped, P.J., Evans, J.J.: Environment feature extraction and classification for context aware physical activity monitoring. In: SAS 2013, pp. 123–128 (2013)Google Scholar
- 16.Kononenko, I., Machine, K.M., Learning, D.M.: Introduction to Principles and Algorithms. Horwood Publishing Limited, Chichester (2007)Google Scholar
- 32.Synnott, J., Chen, L., Nugent, C.D., Moore, G.: The creation of simulated activity datasets using a graphical intelligent environment simulation tool, pp. 4143–4146 (2014)Google Scholar
- 33.Nugent, C., Synnott, J., Santanna, A., Espinilla, M., Cleland, I., Banos, L.J.O., et al.: An initiative for the creation of open datasets within the pervasive healthcare, pp. 180–183 (2016)Google Scholar
- 34.Van Kasteren, T., Noulas, A., Englebienne, G., Krse, B.: Accurate activity recognition in a home setting, pp. 1–9 (2008)Google Scholar
- 35.Alcalá, J., Fernández, A., Luengo, J., Derrac, J., García, S., Sánchez, L.: KEEL data-mining software tool: data set repository, integration of algorithms and experimental analysis framework. J. Multiple-Valued Logic Soft Comput. 17(2–3), 255–287 (2011)Google Scholar