6664 research outputs found
Sort by
Event-based decision support algorithm for real-time flood forecasting in urban drainage systems using machine learning modelling
Get PDFUrban flooding is a major problem for cities around the world, with significant socio-economic consequences.
Conventional real-time flood forecasting models rely on continuous time-series data and often have limited
accuracy, especially for longer lead times than 2 hrs. This study proposes a novel event-based decision support
algorithm for real-time flood forecasting using event-based data identification, event-based dataset generation,
and a real-time decision tree flowchart using machine learning models. The results of applying the framework to
a real-world case study demonstrate higher accuracy in forecasting water level rise, especially for longer lead
times (e.g., 2–3 hrs), compared to traditional models. The proposed framework reduces root mean square error
by 50%, increases accuracy of flood forecasting by 50%, and improves normalised Nash–Sutcliffe error by 20%.
The proposed event-based dataset framework can significantly enhance the accuracy of flood forecasting,
reducing the occurrences of both false alarms and flood missing and improving emergency response systems
“Fancy a Brew? “: Understanding factors influencing ease of use of cups used in care homes
No full textBackground And Aims: There are a wide variety of different designs for mugs and cups, but these are primarily driven by visual aesthetics rather than utility. The range of drinking vessels available to the care home sector is limited and not informed by ergonomic considerations that would make them more suitable for the frail elderly to use. Although our previous work has thrown some light on this problem, there is a need to improve our understanding of the ergonomics of drinking and drinking vessels to better inform both the designs available and purchasing decisions of facilities caring for older people.
Methods: This study was split into two phases, an initial qualitative focus group study and a quantitative ergonomic analysis.
Results: From the focus group study, two cups were preferred of the five presented. The characteristics shared by these two cups were lightness and large handle. From the ergonomic analysis the general grip observed in this to hold a cup can be classified as a power grip with an adducted thumb. Cups with a relatively low mass (m), a handle orifice area (S) sufficient to allow a minimum of two fingers to pass through comfortably whilst offering the ability to be supported by an adducted thumb and ring finger comfortably are seen to perform best. Further, whilst the handle orifice area should be sufficiently large for the optimal grip to be used it should also minimize the moment on the user’s wrist. Computed finger forces show considerable variability across the fingers and across the cups. All the forces calculated from the simulation are relatively low for power grips of the type described earlier. This indicates that the individual finger grip forces are less of an issue for users than the stability needed to control and balance the force in the wrist.
Conclusion: This study has also shown that there are several critical dimensions for the design of cups for people with reduced strength and dexterity. The mass of the cup (m), the diameter of the cup D, the handle length L, and the orifice area S effecting the critical moment on the wrist and the ability to support this moment through the fingers
Editorial: Reflections on a successful year of the Society for Reproductive and Infant Psychology.
Get PDFAs 2023 is drawing to a close, it is timely for us, the committee of the Society for
Reproductive and Infant Psychology (SRIP), to pause and reflect upon what has been
achieved this year and update our Society members and readers of our affiliated journal
‘The Journal of Reproductive and Infant Psychology (JRIP)’ on the plans outlined earlier in
the year (Garthus-Niegel & Horsch, 2023
Sentiment Analysis and Student Emotions: Improving Satisfaction in Online Learning Platforms
No full textHuman emotion recognition using artificial intelligence is among the most prominent research areas. Human-Computer Interaction (HCI) and Sentiment Analysis (SA) are extensively used to detect human emotions. The role of students' sentiments and emotions becomes vital while using online learning platforms due to the need for physical interaction between instructor and learner to understand each other compared to face-to-face learning. This study identifies challenges and issues of online learning and student satisfaction while using online learning platforms. In addition, we analysed the importance of sentiment analysis for students' satisfaction while using online learning platforms. This study retrieved 112 research articles/reports using keywords such as ‘Online Learning’, ‘Online Learning Post-COVID-19’, ‘Student Satisfaction’, ‘Student Sentiment Analysis’, ‘Learning Environments’ and ‘Student Performance in Online Learning’. Among 112 research articles, 37 were identified after excluding duplicate and irrelevant research articles based on defined elimination criteria, such as studies on student emotions related to non-educational purposes. For systematic literature, the study has been divided into two research questions, i.e., (i) exploring key challenges of online learning and (ii) the role of sentiment analysis during online learning. These two questions have been answered based on arguments and evidence from the existing literature
A Transparent CAPTCHAS Verification System for Cloud-Based Smart and Secure Applications
No full textIn times of technological boom, web-based or internet applications use increases daily. Therefore, the need for security is increasing daily due to technological advancements. This paper presents a novel and Transparent CAPTCHA Verification System explicitly designed for cloud-based secure applications. The system generated text from image captchas using advanced deep-learning techniques. Another contribution to this system is that we utilized explainable AI (XAI) to ensure transparency throughout the working process. The generated explanations provide valuable insights into the conversion from image to text, fostering user trust and comprehension. The novel CAPTCHA Verification System has the capacity to significantly enhance the lives of individuals with disabilities (PwD) by providing an inclusive and user-centric approach for verifying their identity across diverse online platforms. By ensuring a transparent and precise transformation of image captchas into text, this system empowers PwD to effortlessly participate in cloud-based applications, enriching their digital interactions and overall well-being. This system has promising applications in various domains, especially in captcha-based verification, where it can significantly improve security and user experience
Adopting Artificial Intelligence in Industry 4.0: Understanding the Drivers, Barriers and Technology Trends
No full textArtificial intelligence (AI) is recognized as an area of strategic importance and a key driver of sustainable development in the era of Industry 4.0. AI is a field of research that deals with the simulation of human intelligence processes using machines and computer systems. Many industries within the manufacturing, energy, construction, aerospace, transport, and healthcare sectors are adopting AI technologies and advanced data analytics tools with the goal to improve their business performance and user experience. Despite some success stories, numerous surveys report that many industries have been slow or failed to adopt AI beyond the proof-of-concept phase to the enterprise scale. This study aims to explore the drivers and barriers to the adoption of AI, and then assess the readiness of industries in implementing AI and big data technologies within their organizations. To gather industry's views on AI and its impact on business models and customer relationships, a close-ended questionnaire is designed and distributed to the respondents. A strategic analysis using the SWOT method is performed to identify the strengths, weaknesses, opportunities, and threats that may face industries during the AI implementation process. The major challenges identified include the lack of access to IT infrastructure and skilled AI talent; lack of good quality data; weak business cases; and complications around policies, regulations, and ethics. To examine the maturity of industries in adopting AI technologies, we used the NASA Technology Readiness Level (TRL) metric, that is based on a scale from 1 to 9 with 1 being the least mature and 9 being the most mature technology. A large majority of respondents have indicated that their TRL in terms of AI infrastructure is between 3 and 4, whereas the TRL in terms of software platforms lies between 3 and 6. Finally, some recommendations are made to help industries overcome challenges in moving to a higher TRL scale
A SWOT Analysis of Pilot Implementation
Get PDFOver the preceding decades, usability testing has become widely used for revealing design problems in information systems, while they are still at the prototype stage. Normally, these tests involve removing users from their work for an hour or two to have them solve pre-set tasks with a system prototype in a lab-like setting. As a result, usability testing is insensitive to many of the organizational and contextual issues that determine the fit between a system and its real-world environment. Methods such as work domain analysis and scenario-based design aim to address this limitation but from an analysis-and-design perspective. In contrast, pilot implementation is an evaluation method. It involves evaluating a system in the field and is, thereby, an important supplement to usability testing.
Evaluation in the field allows for identifying subtle organizational and contextual issues that are critical to the adoption of a system and to its consequences for those affected by it. This makes pilot implementation valuable to the interaction designer. However, pilot implementations are challenging to conduct, the identified issues may be muddled, and the possibilities for resolving them may be limited. In deciding whether and when to apply the method of pilot implementation, interaction designers need to be aware of its strengths, weaknesses, opportunities, and threats (SWOT). In this article, we offer a critical perspective on the adoption of pilot implementation in interaction design, supported by the results of a SWOT analysis
Identifying patients at increased risk of non-ventilator-associated pneumonia on admission to hospital: a pragmatic prognostic screening tool to trigger preventative action
Get PDFBackground
Non-ventilator healthcare-associated pneumonia (NV-HAP) is an important healthcare-associated infection. This study tested the feasibility of using routine admission data to identify those patients at high risk of NV-HAP who could benefit from targeted, preventive interventions.
Methods
Patients aged ≥64 years who developed NV-HAP five days or more after admission to elderly-care wards, were identified by retrospective case note review together with matched controls. Data on potential predictors of NV-HAP were captured from admission records. Multi-variate analysis was used to build a prognostic screening tool (PRHAPs); acceptability and feasibility of the tool was evaluated.
Results
A total of 382 cases/381 control patients were included in the analysis. Ten predictors were included in the final model; nine increased the risk of NV-HAP (OR between 1.68 and 2.42) and one (independent mobility) was protective (OR 0.48; 95% CI 0.30–0.75). The model correctly predicted 68% of the patients with and without NV-HAP; sensitivity 77%; specificity 61%. The PRHAPs tool risk score was 60% or more if two predictors were present and over 70% if three were present. An expert consensus group supported incorporating the PRHAPs tool into electronic logic systems as an efficient mechanism to identify patients at risk of NV-HAP and target preventative strategies.
Conclusions
This prognostic screening (PRHAPs) tool, applied to data routinely collected when a patient is admitted to hospital, could enable staff to identify patients at greatest risk of NV-HAP, target scarce resources in implementing a prevention care bundle, and reduce the use of antimicrobial agents
InSAR for dielectric constant estimation of pavements: A feasibility study.
No full textTransport infrastructure health monitoring is fundamental to assure the road's integrity and the user's safety. Nowadays, pavement quality is conventionally assessed using different Non-Destructive Testing (NDT) methods. These methodologies are adequate but are limited in terms of spatial and temporal coverage. Depth estimation using InSAR data is not uncommon for application in arid areas, glaciers, and soil moisture studies. These areas are subject to substantial penetration at microwave frequencies. The methods assume the medium as a uniform volume with infinite depth and analyze the volume scattering of the radar wave. Determination of dielectric parameters of pavement layers and material inhomogeneities assessment is essential for pavement health monitoring. Here we examine the capability of InSAR data to estimate dielectric constant values in concrete pavements by interaction with the electromagnetic waves of radar satellite data. TerraSAR-X data are used to explore the dielectric constant estimation capabilities. The existing methods to invert for dielectric constant from the InSAR signal are applied to arid, ice-covered, and forested areas, but not pavements. This attempt will be the first of its kind to explore the large area coverage and with a frequent revisit of SAR satellites as a monitoring tool
Improving Nursing Educational Practices and Professional Development through Smart Education in Smart Cities: A Systematic Literature Review
No full textSmart education has gained popularity in modern educational practices, reflecting current nursing education demands for more innovative pedagogical approaches. Smart learning combines student-centred learning, which can be easily accessible in smart cities while at the same time promoting inclusivity and personalisation. A concept similar to current nursing official bodies' requests for future development. The authors of this paper conducted a Systematic Literature Review (SLR), investigating how effective, engaging and user-friendly current computer-aided nursing pedagogies are. Moreover, this SLR examines current computer-aided nursing pedagogies' accessibility in smart education and smart cities. This paper reviews 53 primary research articles, including Randomised Control Trials (RCTs), Quasi-experimental studies, qualitative and mixed-methods observational studies, conference papers and pilot studies. The overall findings of this SLR suggest the overall effectiveness and user-friendliness of computer-aided nursing education pedagogies. Furthermore, we found that concepts such as gamification and systems with interactive features were most beneficial in attracting nursing students' engagement