An evolving machine learning method for human activity recognition systems

Andréu, Javier; Angelov, Plamen

doi:10.1007/s12652-011-0068-9

Cited by 22 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experimental results compare favourably with other work using body sensors (Stefan et al, 2012). Moreover, the performance of our approach shows a significant improvement in comparison to the approach using a single smartphone (Zhang et al, 2010) and the approach built on fixed machine learning algorithms (Andreu and Angelov, 2013). This shows a lot of promise for using smartphones as an alternative to dedicated sensors and using the cloud-based data analytics framework to process machine learning tasks.…”

Section: Discussionmentioning

confidence: 84%

Smartphone-based Activity Recognition using Hybrid Classifier - Utilizing Cloud Infrastructure for Data Analysis

Yuan¹,

Herbert²,

Emamian³

2014

Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems

View full text Add to dashboard Cite

Learning and recognizing the activities of daily living (ADLs) of an individual is vital when providing an individual with context-aware at-home healthcare. In this work, unobtrusive detection of inhabitants' activities in the home environment is implemented through the smartphone and wearable wireless sensor belt solution. A hybrid classifier is developed by combining threshold-based methods and machine learning mechanisms. Features extracted from the raw inertial sensor data are collected from a Body Area Network (BAN) (consisting of the Zephyr BioHarness sensor and an Android smartphone), and are used to build classification models using different machine learning algorithms. A cloud-based data analytics framework is developed to process different classification models in parallel and to select the most suitable model for each user. The evolving machine learning mechanism makes the model become customizable and self-adaptive by utilizing a cloud infrastructure which also overcomes the limitation of the computing power and storage of a smartphone. Furthermore, we investigate methods for adapting a universal model, which is trained using the data set of all users, to an individual user through an unsupervised learning scheme. The evaluation results of the experiments conducted on eight participants indicate that the proposed approach can robustly identify activities in real-time across multiple individuals: the highest recognition rate achieved 98% after a few runs.

show abstract

Section: Discussionmentioning

confidence: 84%

Smartphone-based Activity Recognition using Hybrid Classifier - Utilizing Cloud Infrastructure for Data Analysis

Yuan¹,

Herbert²,

Emamian³

2014

Proceedings of the 4th International Conference on Pervasive and Embedded Computing and Communication Systems

View full text Add to dashboard Cite

show abstract

“…Some data stream algorithms have been implemented considering adaptation such as adaptive, very fast decision tree learners [105], which has been used for sleep apnoea monitoring [106]. Likewise, an adaptive neuro-fuzzy stream learning approach was proposed for the recognition of activities of daily living [107,108]. Additionally, signal processing algorithms such as symbolic aggregate approximation can also implement adaptation to deal with data stream segmentation and approximation [109].…”

Section: H From Sensor Informatics To Big Datamentioning

confidence: 99%

From Wearable Sensors to Smart Implants-–Toward Pervasive and Personalized Healthcare

Andreu-Pérez

Leff

et al. 2015

IEEE Trans. Biomed. Eng.

299

156

View full text Add to dashboard Cite

Abstract-Objective:This article discusses the evolution of pervasive healthcare from its inception for activity recognition using wearable sensors to the future of sensing implant deployment and data processing. Methods: We provide an overview of some of the past milestones and recent developments, categorised into different generations of pervasive sensing applications for health monitoring. This is followed by a review on recent technological advances that have allowed unobtrusive continuous sensing combined with diverse technologies to reshape the clinical workflow for both acute and chronic disease management. We discuss the opportunities of pervasive health monitoring through data linkages with other health informatics systems including the mining of health records, clinical trial databases, multi-omics data integration and social media. Conclusion: Technical advances have supported the evolution of the pervasive health paradigm towards preventative, predictive, personalised and participatory medicine. Significance: The sensing technologies discussed in this paper and their future evolution will play a key role in realising the goal of sustainable healthcare systems.Index Terms-pervasive health, wearable sensors, implantable sensors, health informatics.

show abstract

“…Due to their evolving and resilient characters, ML classifiers have been implemented in a variety of applications built on WSANs (Alsheikh et al 2014). HAR, as one such application, has successfully exploited classifiers in the last five years (see, for example, (Xiao and Lu 2015;Villa et al 2012;Andreu and Angelov 2013). However, due to the cost-effective and low energy-consumption character typical of WSAN nodes, computational processing with respect to feature extraction has been considerably limited (Salomons et al 2016).…”

Section: Machine Learning 10mentioning

confidence: 99%

Visions of Process—Swarm Intelligence and Swarm Robotics in Architectural Design and Construction

Vehlken

2018

Springer Series in Adaptive Environments

View full text Add to dashboard Cite

Robotic Building implies both physically built robotic environments and robotically supported building processes. Physically built robotic environments consist of reconfigurable, adaptive systems incorporating sensor-actuator mechanisms that enable buildings to interact with their users and surroundings in real-time. These robotic environments require Design-to-Production and-Operation (D2P&O) chains that may be (partially or completely) robotically driven. This chapter describes previous work aiming to integrate D2RP&O processes by linking performance-driven design with robotic production and user-driven building operation. 5.1 Introduction While architecture and architectural production are increasingly incorporating aspects of non-human agency employing data, information, and knowledge contained within the (worldwide) network connecting electronic devices, the question is not whether but how robotic systems can be incorporated into building processes and buildings (Oosterhuis and Bier 2013). This chapter aims to answer this question by reflecting on the achievements of the Robotic Building (RB) team at Technical University Delft (TU Delft) and by identifying future steps. The focus is on an architecture that is robotically enabled to interact with its users and surroundings in real-time and the corresponding Design-to-Production and-Operation (D2P&O) processes that are (in part or as whole) robotically driven. Such modes of production and operation involve agency of both humans and non-humans. Thus agency is not The original version of the book was revised: Open access text has been updated in FM, Chapter

show abstract

An evolving machine learning method for human activity recognition systems

Cited by 22 publications

References 13 publications

Smartphone-based Activity Recognition using Hybrid Classifier - Utilizing Cloud Infrastructure for Data Analysis

Smartphone-based Activity Recognition using Hybrid Classifier - Utilizing Cloud Infrastructure for Data Analysis

From Wearable Sensors to Smart Implants-–Toward Pervasive and Personalized Healthcare

Visions of Process—Swarm Intelligence and Swarm Robotics in Architectural Design and Construction

Contact Info

Product

Resources

About