scholarly journals EAGLE—A Scalable Query Processing Engine for Linked Sensor Data

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4362 ◽  
Author(s):  
Hoan Nguyen Mau Quoc ◽  
Martin Serrano ◽  
Han Mau Nguyen ◽  
John G. Breslin ◽  
Danh Le-Phuoc
Keyword(s):  
Poor Performance ◽  
Heterogeneous Data ◽  
Sensor Data ◽  
Semantic Web Technologies ◽  
Heterogeneous Data Integration ◽  
Temporal Correlations ◽  
Temporal Query ◽  
Query Engine ◽  
Aggregate Queries ◽  
Spatio Temporal

Recently, many approaches have been proposed to manage sensor data using semantic web technologies for effective heterogeneous data integration. However, our empirical observations revealed that these solutions primarily focused on semantic relationships and unfortunately paid less attention to spatio–temporal correlations. Most semantic approaches do not have spatio–temporal support. Some of them have attempted to provide full spatio–temporal support, but have poor performance for complex spatio–temporal aggregate queries. In addition, while the volume of sensor data is rapidly growing, the challenge of querying and managing the massive volumes of data generated by sensing devices still remains unsolved. In this article, we introduce EAGLE, a spatio–temporal query engine for querying sensor data based on the linked data model. The ultimate goal of EAGLE is to provide an elastic and scalable system which allows fast searching and analysis with respect to the relationships of space, time and semantics in sensor data. We also extend SPARQL with a set of new query operators in order to support spatio–temporal computing in the linked sensor data context.

VGEs-Oriented Multi-sourced Heterogeneous Data Integration

2010 ◽  
Vol 11 (3) ◽  
pp. 292-298
Author(s):  
Hongjun SU ◽  
Yehua SHENG ◽  
Yongning WEN ◽  
Min CHEN
Keyword(s):  
Data Integration ◽  
Heterogeneous Data ◽  
Heterogeneous Data Integration

scholarly journals Interpretable deep learning for the remote characterisation of ambulation in multiple sclerosis using smartphones

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Andrew P. Creagh ◽  
Florian Lipsmeier ◽  
Michael Lindemann ◽  
Maarten De Vos
Keyword(s):  
Multiple Sclerosis ◽  
Deep Learning ◽  
Inertial Sensor ◽  
Heterogeneous Data ◽  
Fine Tuning ◽  
Sensor Data ◽  
Support Vector ◽  
Deep Convolutional Neural Networks ◽  
Healthcare Applications ◽  
Feature Based

AbstractThe emergence of digital technologies such as smartphones in healthcare applications have demonstrated the possibility of developing rich, continuous, and objective measures of multiple sclerosis (MS) disability that can be administered remotely and out-of-clinic. Deep Convolutional Neural Networks (DCNN) may capture a richer representation of healthy and MS-related ambulatory characteristics from the raw smartphone-based inertial sensor data than standard feature-based methodologies. To overcome the typical limitations associated with remotely generated health data, such as low subject numbers, sparsity, and heterogeneous data, a transfer learning (TL) model from similar large open-source datasets was proposed. Our TL framework leveraged the ambulatory information learned on human activity recognition (HAR) tasks collected from wearable smartphone sensor data. It was demonstrated that fine-tuning TL DCNN HAR models towards MS disease recognition tasks outperformed previous Support Vector Machine (SVM) feature-based methods, as well as DCNN models trained end-to-end, by upwards of 8–15%. A lack of transparency of “black-box” deep networks remains one of the largest stumbling blocks to the wider acceptance of deep learning for clinical applications. Ensuing work therefore aimed to visualise DCNN decisions attributed by relevance heatmaps using Layer-Wise Relevance Propagation (LRP). Through the LRP framework, the patterns captured from smartphone-based inertial sensor data that were reflective of those who are healthy versus people with MS (PwMS) could begin to be established and understood. Interpretations suggested that cadence-based measures, gait speed, and ambulation-related signal perturbations were distinct characteristics that distinguished MS disability from healthy participants. Robust and interpretable outcomes, generated from high-frequency out-of-clinic assessments, could greatly augment the current in-clinic assessment picture for PwMS, to inform better disease management techniques, and enable the development of better therapeutic interventions.

scholarly journals Stereo-PIV measurements of spatio-temporal turbulence correlations in an axisymmetric jet

2015 ◽  
Vol 778 ◽  
pp. 216-252 ◽  
Author(s):  
C. D. Pokora ◽  
J. J. McGuirk
Keyword(s):  
Fourth Order ◽  
Single Point ◽  
Aerodynamic Noise ◽  
Filter Method ◽  
Frequency Noise ◽  
Turbulent Structures ◽  
Validation Data ◽  
Piv Measurements ◽  
Temporal Correlations ◽  
Spatio Temporal

Stereoscopic three-component particle image velocimetry (3C-PIV) measurements have been made in a turbulent round jet to investigate the spatio-temporal correlations that are the origin of aerodynamic noise. Restricting attention to subsonic, isothermal jets, measurements were taken in a water flow experiment where, for the same Reynolds number and nozzle size, the shortest time scale of the dynamically important turbulent structures is more than an order of magnitude greater that in equivalent airflow experiments, greatly facilitating time-resolved PIV measurements. Results obtained (for a jet nozzle diameter and velocity of 40 mm and $1~\text{m}~\text{s}^{-1}$, giving $\mathit{Re}=4\times 10^{4}$) show that, on the basis of both single-point statistics and two-point quantities (correlation functions, integral length scales) the present incompressible flow data are in excellent agreement with published compressible, subsonic airflow measurements. The 3C-PIV data are first compared to higher-spatial-resolution 2C-PIV data and observed to be in good agreement, although some deterioration in quality for higher-order correlations caused by high-frequency noise in the 3C-PIV data is noted. A filter method to correct for this is proposed, based on proper orthogonal decomposition (POD) of the 3C-PIV data. The corrected data are then used to construct correlation maps at the second- and fourth-order level for all velocity components. The present data are in accordance with existing hot-wire measurements, but provide significantly more detailed information on correlation components than has previously been available. The measured relative magnitudes of various components of the two-point fourth-order turbulence correlation coefficient ($R_{ij,kl}$) – the fundamental building block for free shear flow aerodynamic noise sources – are presented and represent a valuable source of validation data for acoustic source modelling. The relationship between fourth-order and second-order velocity correlations is also examined, based on an assumption of a quasi-Gaussian nearly normal p.d.f. for the velocity fluctuations. The present results indicate that this approximation shows reasonable agreement for the measured relative magnitudes of several correlation components; however, areas of discrepancy are identified, indicating the need for work on alternative models such as the shell turbulence concept of Afsar (Eur. J. Mech. (B/Fluids), vol. 31, 2012, pp. 129–139).

Sign in / Sign up

Export Citation Format

Share Document