Novel evolutionary-EAC instance-learning-based algorithm for fast data stream mining in assisted living with extreme connectivity


In modern healthcare, sensing technologies such as IoT empower the quality of assisted living service by knowing what a resident is doing in real-time. Using extreme connectivity and cloud computing in a smart home, where a collection of sensors is installed, the sensors sample continuously from the movements of the resident as well as ambient data from the surrounding inside the house. Automatic human activity recognition of the resident's activities is one of the key components of assisted living in smart home. For monitoring in-home safety, the ability in recognizing abnormal activities such as accident, falling, acute disease attack (e.g. asthma, stroke, etc.), fainting, wobbling, is particularly important. The detection and machine learning process must be both accurate and fast, to cope with the real-time activity recognition. To this end, a novel streamlined sensor data processing method is proposed called Evolutionary Expand-and-Contract Instance-based Learning algorithm (EEAC-IBL). The multivariate data stream is first expanded into many subspaces, then the subspaces which are corresponding to the characteristics of the features are selected and condensed into a significant feature subset. The selection operates scholastically instead of deterministically by evolutionary optimization which approximates the best subgroup. Followed by data stream mining, the machine learning for activity recognition is done on the fly. This approach is unique and suitable for such extreme connectivity scenario where precise feature selection is not required, and the relative importance of each feature among the sensor data changes over time. This stochastic approximation method is fast and accurate, offering an alternative to traditional machine learning method for smart home activity recognition application. Our experimental results show computing advantages over other classical approaches.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10


  1. 1.

    Roy N, Misra A, Cook D (2016) Ambient and smartphone sensor assisted ADL recognition in multi-inhabitant smart environments. J Ambient Intell Human Comput 7:1–19.

    Article  Google Scholar 

  2. 2.

    Sriram RD, Sheth A (2015) Internet of things perspectives. IT Prof 17(3):60–63

    Article  Google Scholar 

  3. 3.

    Mizuno H, Nagai H, Sasaki K, Hosaka H, Sugimoto C, Khalil K, Tatsuta S (2007) Wearable sensor system for human behavior recognition (first report: basic architecture and behavior prediction method). In: TRANSDUCERS 2007—2007 international solid-state sensors, actuators and microsystems conference, pp 435–438

  4. 4.

    Jalal A, Kim Y, Kamal S, Farooq A, Kim D (2015) Human daily activity recognition with joints plus body features representation using Kinect sensor. In: 2015 international conference on informatics, electronics and vision (ICIEV), pp 1–6

  5. 5.

    Ahad MAR, Antar AD, Ahmed M, IoT sensor-based activity recognition. Springer, ISBN 978–3–030–51378–8

  6. 6.

    Bilal M, Shaikh FK, Arif M et al (2019) A revised framework of machine learning application for optimal activity recognition. Cluster Comput 22:7257–7273.

    Article  Google Scholar 

  7. 7.

    Suto J, Oniga S, Lung C et al (2018) Comparison of offline and real-time human activity recognition results using machine learning techniques. Neural Comput Appl.

    Article  Google Scholar 

  8. 8.

    Dash Y, Kumar S, Patle VK (2016) A novel data mining scheme for smartphone activity recognition by accelerometer sensor. In: Das S, Pal T, Kar S, Satapathy S, Mandal J (eds) Proceedings of the 4th international conference on frontiers in intelligent computing: theory and applications (FICTA) 2015. Advances in intelligent systems and computing, vol 404. Springer, New Delhi.

  9. 9.

    Li F, Shirahama K, Nisar MA, Huang X, Grzegorzek M (2020) Deep transfer learning for time series data based on sensor modality classification. Sensors 20:4271

    Article  Google Scholar 

  10. 10.

    Sawada Y, Sato Y, Nakada T, Yamaguchi S, Ujimoto K, Hayashi N (2019) Improvement in classification performance based on target vector modification for all-transfer deep learning. Appl Sci 9:128

    Article  Google Scholar 

  11. 11.

    Ding R, Li X, Nie L, Li J, Si X, Chu D, Liu G, Zhan D (2019) Empirical study and improvement on deep transfer learning for human activity recognition. Sensors 19:57

    Article  Google Scholar 

  12. 12.

    Boukli Hacene G, Gripon V, Farrugia N, Arzel M, Jezequel M (2018) Transfer incremental learning using data augmentation. Appl Sci 8:2512

    Article  Google Scholar 

  13. 13.

    Sajjad M, Zahir S, Ullah A et al (2020) Human behavior understanding in big multimedia data using CNN based facial expression recognition. Mobile Netw Appl 25:1611–1621.

    Article  Google Scholar 

  14. 14.

    Vandersmissen B, Knudde N, Jalalvand A et al (2020) Indoor human activity recognition using high-dimensional sensors and deep neural networks. Neural Comput Appl 32:12295–12309.

    Article  Google Scholar 

  15. 15.

    Fong S, Liu K, Cho K et al (2016) Improvised methods for tackling big data stream mining challenges: case study of human activity recognition. J Supercomput 72:3927–3959.

    Article  Google Scholar 

  16. 16.

    Fong S et al (2018) Performance evaluation of shadow features as a data preprocessing method in data mining for human activities recognitions. In: Wong R, Chi CH, Hung P (eds) Behavior engineering and applications. International series on computer entertainment and media technology. Springer, Cham.

    Chapter  Google Scholar 

  17. 17.

    Ahlawat K, Singh AP (2020) Human activity recognition in imbalanced big data using fuzzy rule-based classification system. In: Reddy V, Prasad V, Wang J, Reddy K (eds) Soft computing and signal processing. ICSCSP 2019. Advances in intelligent systems and computing, vol 1118. Springer, Singapore.

    Chapter  Google Scholar 

  18. 18.

    Amezzane I, Fakhri Y, El Aroussi M, Bakhouya M (2018) Analysis and effect of feature selection over smartphone-based dataset for human activity recognition. In: Belqasmi F, Harroud H, Agueh M, Dssouli R, Kamoun F (eds) Emerging technologies for developing countries. AFRICATEK 2017. Lecture notes of the institute for computer sciences, social informatics and telecommunications engineering, vol 206. Springer, Cham.

    Chapter  Google Scholar 

  19. 19.

    Gopalakrishnan N, Krishnan V, Gopalakrishnan V (2020) Ensemble feature selection to improve classification accuracy in human activity recognition. In: Ranganathan G, Chen J, Rocha Á (eds) Inventive communication and computational technologies. Lecture notes in networks and systems, vol 89. Springer, Singapore.

    Chapter  Google Scholar 

  20. 20.

    Guha S, Koudas N (2002) Approximating a data stream for querying and estimation: algorithms and performance evaluation. In: Proceedings 18th international conference on data engineering, San Jose, CA, USA, pp 567–576.

  21. 21.

    Lee S-M, Yoon SM, Cho H (2017) Human activity recognition from accelerometer data using convolutional neural network. In: 2017 IEEE international conference on big data and smart computing (BigComp), Jeju, pp 131–134.

  22. 22.

    Babiker M, Khalifa OO, Htike KK, Hassan A, Zaharadeen M (2017) Automated daily human activity recognition for video surveillance using neural network. In: 2017 IEEE 4th international conference on smart instrumentation, measurement and application (ICSIMA), Putrajaya, pp 1–5.

  23. 23.

    Goel A, Abubakr A, Koperski M, Bremond F, Francesca G (2018) Online temporal detection of daily-living human activities in long untrimmed video streams. In: 2018 IEEE international conference on image processing, applications and systems (IPAS), Sophia Antipolis, France, pp 43–48.

  24. 24.

    Jianqin Y, Guohui T, Xinran W (2012) Human activity recognition based on event histogram and KL transform. In: Proceedings of the 31st Chinese control conference, Hefei, pp 3912–3916

  25. 25.

    Yang J, Cheng J, Lu H (2009) Human activity recognition based on the blob features. In: 2009 IEEE international conference on multimedia and expo, New York, NY, pp 358–361.

  26. 26.

    Zhang Y, An H, Ma H, Wei Q, Wang J (2018) Human activity recognition with discrete cosine transform in lower extremity exoskeleton. In: 2018 IEEE international conference on intelligence and safety for robotics (ISR), Shenyang, pp 309–312.

  27. 27.

    Pathan NS, Talukdar MTF, Quamruzzaman M, Fattah SA (2019) A machine learning based human activity recognition during physical exercise using wavelet packet transform of PPG and inertial sensors data. In: 2019 4th international conference on electrical information and communication technology (EICT), Khulna, Bangladesh, pp 1–5.

  28. 28.

    He Z (2010) Activity recognition from accelerometer signals based on Wavelet-AR model. In: 2010 IEEE international conference on progress in informatics and computing, Shanghai, pp 499–502.

  29. 29.

    Tian Y, Wang X, Yang P, Wang J, Zhang J (2018) A single accelerometer-based robust human activity recognition via wavelet features and ensemble feature selection. In: 2018 24th international conference on automation and computing (ICAC), Newcastle upon Tyne, UK, pp 1–6.

  30. 30.

    Zhou ZH (2007) Mining ambiguous data with multi-instance multi-label representation. In: Alhajj R, Gao H, Li J, Li X, Zaïane OR (eds) Advanced data mining and applications. ADMA 2007. Lecture notes in computer science, vol 4632. Springer, Berlin.

    Chapter  Google Scholar 

  31. 31.

    Lange S, Watson P (1994) Machine discovery in the presence of incomplete or ambiguous data. In: Arikawa S, Jantke KP (eds) Algorithmic learning theory. AII 1994, ALT 1994. Lecture notes in computer science (lecture notes in artificial intelligence), vol 872. Springer, Berlin.

    Chapter  Google Scholar 

  32. 32.

    Babenko B (2008) Multiple instance learning: algorithms and applications. View Article PubMed/NCBI Google Scholar

  33. 33.

    Guindon S, Gascuel O (2019) Numerical optimization techniques in maximum likelihood tree inference. In: Warnow T (ed) Bioinformatics and phylogenetics. Computational biology, vol 29. Springer, Cham.

    Chapter  Google Scholar 

  34. 34.

    Wu X, Kumar V, Quinlan JR, Ghosh J, Yang Q, Motoda H, McLachlan GJ, Ng A, Liu B, Philip SY, Zhou ZH (2008) Top 10 algorithms in data mining. Knowl Inf Syst 14(1):1–37.

    Article  Google Scholar 

  35. 35.

    Cook D (2012) Learning setting-generalized activity models for smart spaces. IEEE Intell Syst 27(1):32–38

    Article  Google Scholar 

  36. 36.

    Bifet A, Holmes G, Kirkby R, Pfahringer B (2010) MOA: massive online analysis. J Mach Learn Res 11:1601–1604

    Google Scholar 

  37. 37.

    Domingos P, Hulten G (2000) Mining high-speed data streams. In: Proceedings KDD 2000. ACM Press, New York, pp 71–80

  38. 38.

    Yang H, Fong S (2011) Optimized very fast decision tree with balanced classification accuracy and compact tree size. In: The 3rd international conference on data mining and intelligent information technology applications, Macao, pp 57–64

  39. 39.

    Fong S, Biuk-Aghai RP, Millham RC (2018) Swarm search methods in weka for data mining. In: ICMLC 2018: proceedings of the 2018 10th international conference on machine learning and computing, pp 122–127

  40. 40.

    Piao Y, Ryu KH (2017) A hybrid feature selection method based on symmetrical uncertainty and support vector machine for high-dimensional data classification. In: Nguyen N, Tojo S, Nguyen L, Trawiński B (eds) Intelligent information and database systems. ACIIDS 2017. Lecture notes in computer science, vol 10191. Springer, Cham.

    Chapter  Google Scholar 

  41. 41.

    Fong S, Wong R, Vasilakos AV (2016) Accelerated PSO swarm search feature selection for data stream mining big data. IEEE Trans Serv Comput 9(1):33–45.

    Article  Google Scholar 

  42. 42.

    Fong S, Liang J, Siu SW, Chan JH (2015) Efficient variation-based feature selection for medical data classification. J Med Imaging Health Inf 5(5):1093–1098

    Article  Google Scholar 

Download references


The authors are thankful for the financial support from the research grants, MYRG2016-00069, entitled ’Nature-Inspired Computing and Metaheuristics Algorithms for Optimizing Data stream mining Performance’, EF003/FST-FSJ/2019/GSTIC, code no. 201907010001, FDCT/126/2014/A3, entitled ‘A Scalable Data Stream Mining Methodology: Stream-based Holistic Analytics and Reasoning in Parallel’ offered by FDCT and RDAO/FST, the University of Macau and the Macau SAR government.

Author information




Conceptualization, S. H. and S. F.; Data curation, S. H.; Investigation, R. C. M.; Methodology, K.C. and S. F; Resources, S. H., R. C. M. and S. F.; Software, S. H. and S. F.; Supervision, J. F.; Validation, J. F. and K. C.; Visualization, W. S.; Writing – original draft, S. H.; Writing – review & editing, S. F., R. C. M, W. S., J. F. and K. C. All authors have read and agreed to the published version of the manuscript.

Corresponding author

Correspondence to Simon Fong.

Ethics declarations

Conflict of interest

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hu, S., Fong, S., Song, W. et al. Novel evolutionary-EAC instance-learning-based algorithm for fast data stream mining in assisted living with extreme connectivity. Computing 103, 1519–1543 (2021).

Download citation


  • Human activity recognition
  • IoT data analysis
  • Forecasting
  • Regression
  • Assisted living
  • Extreme connectivity

Mathematics Subject Classification

  • 68W50