scholarly journals Homogeneous Data Normalization and Deep Learning: A Case Study in Human Activity Classification

2020 ◽  
Vol 12 (11) ◽  
pp. 194
Author(s):  
Ivan Miguel Pires ◽  
Faisal Hussain ◽  
Nuno M. M. Garcia ◽  
Petre Lameski ◽  
Eftim Zdravevski
Keyword(s):  
Mobile Devices ◽  
Human Activity ◽  
Human Activities ◽  
Data Normalization ◽  
Accelerometer Data ◽  
Classification Problems ◽  
Intelligent Methods ◽  
The Impact ◽  
Homogeneous Data

One class of applications for human activity recognition methods is found in mobile devices for monitoring older adults and people with special needs. Recently, many studies were performed to create intelligent methods for the recognition of human activities. However, the different mobile devices in the market acquire the data from sensors at different frequencies. This paper focuses on implementing four data normalization techniques, i.e., MaxAbsScaler, MinMaxScaler, RobustScaler, and Z-Score. Subsequently, we evaluate the impact of the normalization algorithms with deep neural networks (DNN) for the classification of the human activities. The impact of the data normalization was counterintuitive, resulting in a degradation of performance. Namely, when using the accelerometer data, the accuracy dropped from about 79% to only 53% for the best normalization approach. Similarly, for the gyroscope data, the accuracy without normalization was about 81.5%, whereas with the best normalization, it was only 60%. It can be concluded that data normalization techniques are not helpful in classification problems with homogeneous data.

scholarly journals Development of transport infrastructure as precondition of economy and tourism development: Case study municipality of Negotin

2013 ◽  
Vol 93 (3) ◽  
pp. 145-157
Author(s):  
Nikola Ristic ◽  
Bogdan Lukic ◽  
Dejan Filipovic ◽  
Velimir Secerov
Keyword(s):  
Functional Connectivity ◽  
Human Activities ◽  
Tourism Development ◽  
Transport Network ◽  
Transport Infrastructure ◽  
Mutual Interaction ◽  
The Impact

Developed transport network is a precondition for economic and tourism development of areas and largely follows and allows the development of human activities. If it is developing without plan, spontaneous and without coordination it may be a limit to the overall development. The aim of research was to define developmental basis for the revitalization, improvement and construction of transport infrastructure in the municipality of Negotin. The paper will present the mutual interaction and functional connectivity of planning solutions for development of transport infrastructure and development of economic and tourism, as well as the impact which planning solutions have on the evolvent of other spatial and city functions.

Supporting Mobile Access to VLE Resources through MobiGlam

2012 ◽  
pp. 315-332
Author(s):  
Fatma Meawad ◽  
Geneen Stubbs
Keyword(s):  
Learning Environments ◽  
Mobile Devices ◽  
Virtual Learning Environments ◽  
Added Value ◽  
Learning Activity ◽  
Mobile Access ◽  
Generic Framework ◽  
The University ◽  
The Impact

MobiGlam is a generic framework of interoperability with existing virtual learning environments (VLEs) that provides a compact and easy to use implementation of learning activity on Java enabled mobile devices. A case study was conducted at the University of Glamorgan, UK where MobiGlam was seamlessly integrated with the university’s VLE to support the delivery of computer courses at the foundation level. Such integration showed an added value to the participants and in many cases, it improved their use of the VLE. This chapter reports on the deployment, the evaluation, and the results of this case study. The results are analysed from two views: the impact on the participants’ use of the VLE and the framework’s overall usability.

Applying Mobile Technologies to Banking Business Processes

2011 ◽  
pp. 778-792
Author(s):  
Dinesh Arunatileka
Keyword(s):  
Service Delivery ◽  
Mobile Devices ◽  
Major Part ◽  
Business Processes ◽  
Mobile Technologies ◽  
Battery Life ◽  
Screen Size ◽  
Banking Business ◽  
The Impact

This chapter discusses the impact of mobile technologies on service delivery processes in a banking environment. Advances in mobile technologies have opened up numerous possibilities for businesses to expand their reach beyond the traditional Internet-based connectivity and, at the same time, have created unique challenges. Security concerns, as well as hurdles of delivering mobile services “anywhere and anytime” using current mobile devices with their limitations of bandwidth, screen size and battery life are examples of such challenges. Banks are typically affected by these advances as a major part of their business deals with providing services that can benefit immensely by adoption of mobile technologies. As an example case study, this chapter investigates some business processes of a leading Australian bank in the context of application of mobile technologies.

Reflections on developing Technology-Enhanced Learning through staff-student partnership - a case study of using mobile devices cross-institutionally

2017 ◽  
Vol 3 (1) ◽  
pp. 243 ◽  
Author(s):  
Matt Elphick ◽  
Stuart Sims
Keyword(s):  
Mobile Devices ◽  
Technology Enhanced Learning ◽  
Learning And Teaching ◽  
Pilot Scheme ◽  
Integration Of Technology ◽  
Student Projects ◽  
The University ◽  
The Impact ◽  
Enhanced Learning

Drawing upon project outputs from seven staff-student partnership projects, this case study explores the impact of a pilot programme to integrate the use of mobile devices into learning and teaching at the University of Winchester. This ‘iPilot’ was designed to give students and staff the opportunity to lead change around the integration of technology into teaching, supported by the Student Fellows Scheme (SFS). We outline the principles behind these partnerships and explore the role that having Student Fellows in a pedagogical leadership position had upon the wider project. This article represents the perspectives of both the co-ordinator of the pilot scheme and the SFS to give a centralised view of a project that was devolved to different programmes. While all of the staff-student projects had a degree of success in furthering the way that mobile devices are used in their respective programmes, many projects veered away from the principles of partnership working which were built into the initial plans for the iPilot. We reflect on barriers encountered in this project and make recommendations based on this experience of how to ensure that the key principles of enhancement are being adhered to, rather than using partnership working in a tokenistic way.

scholarly journals Energy-Efficient Real-Time Human Activity Recognition on Smart Mobile Devices

2016 ◽  
Vol 2016 ◽  
pp. 1-12 ◽  
Author(s):  
Jin Lee ◽  
Jungsun Kim
Keyword(s):  
Real Time ◽  
Mobile Devices ◽  
Activity Recognition ◽  
Human Activity ◽  
Energy Efficient ◽  
Window Size ◽  
Human Activity Recognition ◽  
Smart Mobile ◽  
The Impact ◽  
Smart Mobile Devices

Nowadays, human activity recognition (HAR) plays an important role in wellness-care and context-aware systems. Human activities can be recognized in real-time by using sensory data collected from various sensors built in smart mobile devices. Recent studies have focused on HAR that is solely based on triaxial accelerometers, which is the most energy-efficient approach. However, such HAR approaches are still energy-inefficient because the accelerometer is required to run without stopping so that the physical activity of a user can be recognized in real-time. In this paper, we propose a novel approach for HAR process that controls the activity recognition duration for energy-efficient HAR. We investigated the impact of varying the acceleration-sampling frequency and window size for HAR by using the variable activity recognition duration (VARD) strategy. We implemented our approach by using an Android platform and evaluated its performance in terms of energy efficiency and accuracy. The experimental results showed that our approach reduced energy consumption by a minimum of about 44.23% and maximum of about 78.85% compared to conventional HAR without sacrificing accuracy.

scholarly journals Assessing the Country-Level Excess All-Cause Mortality and the Impacts of Air Pollution and Human Activity during the COVID-19 Epidemic

2021 ◽  
Vol 18 (13) ◽  
pp. 6883
Author(s):  
Yuan Meng ◽  
Man Sing Wong ◽  
Hanfa Xing ◽  
Mei-Po Kwan ◽  
Rui Zhu
Keyword(s):  
Air Pollution ◽  
Human Activity ◽  
Human Activities ◽  
Ground Level ◽  
Global Scale ◽  
Country Level ◽  
Relative Risks ◽  
All Cause Mortality ◽  
Specific Mortality ◽  
The Impact

The impact of Coronavirus Disease 2019 (COVID-19) on cause-specific mortality has been investigated on a global scale. However, less is known about the excess all-cause mortality and air pollution-human activity responses. This study estimated the weekly excess all-cause mortality during COVID-19 and evaluated the impacts of air pollution and human activities on mortality variations during the 10th to 52nd weeks of 2020 among sixteen countries. A SARIMA model was adopted to estimate the mortality benchmark based on short-term mortality during 2015–2019 and calculate excess mortality. A quasi-likelihood Poisson-based GAM model was further applied for air pollution/human activity response evaluation, namely ground-level NO2 and PM2.5 and the visit frequencies of parks and workplaces. The findings showed that, compared with COVID-19 mortality (i.e., cause-specific mortality), excess all-cause mortality changed from −26.52% to 373.60% during the 10th to 52nd weeks across the sixteen countries examined, revealing higher excess all-cause mortality than COVID-19 mortality in most countries. For the impact of air pollution and human activities, the average country-level relative risk showed that one unit increase in weekly NO2, PM2.5, park visits and workplace visits was associated with approximately 1.54% increase and 0.19%, 0.23%, and 0.23% decrease in excess all-cause mortality, respectively. Moreover, compared with the impact on COVID-19 mortality, the relative risks of weekly NO2 and PM2.5 were lower, and the relative risks of weekly park and workplace visits were higher for excess all-cause mortality. These results suggest that the estimation based on excess all-cause mortality reduced the potential impact of air pollution and enhanced the influence of human activities compared with the estimation based on COVID-19 mortality.

Sign in / Sign up

Export Citation Format

Share Document