Search

In advanced missions, handling thin and soft membranes is difficult for robots. This task is composed of several sub-tasks, including turning and removing a sheet from a pile of papers on a table, folding paper, and sticking two pieces of paper together. In this study, a hand robot turns and removes a sheet from a pile of papers, and ensures that only one sheet is grasped. To perform this task, two robotic fingers compress the piled papers in the normal direction of the table with a specific force as they close. We therefore incorporated position-based force control into our system; for the force controller, we adopted stiffness control that calculates position modification proportionally to the difference between the target force and an external force measured by the tactile sensor. With this controller, the robot was able to remove a sheet from a pile of papers and ensure that only one sheet was removed. Using this discrimination, the maximum-minimum shearing forces are better classified than with normal force distribution deviation. [ABSTRACT FROM AUTHOR]

Abstract: Use of the Internet in research is a relatively new phenomenon offering a potentially valuable research resource that, although increasingly used, appears largely untapped in nursing and healthcare more generally. This paper discusses methodological and ethical issues that need consideration when designing an Internet-based study concluding that, in general, online research methods are simply adaptations of traditional methods of data collection. Issues such as the representativeness of the data and ethical concerns are discussed. It considers whether the ethical dilemmas faced by online researchers differ from those faced by those seeking to use other, more ‘traditional’ approaches. Using the example of a study that employed the Internet as a means of distributing questionnaires, this paper shows that this can be an efficient and effective means of gathering data from a geographically dispersed sample. Furthermore, since typewritten data is obtained in the same format from all respondents, the need for transcription and the potential for error are reduced potentially enhancing the quality of any such study. [Copyright &y& Elsevier]

The proposal of Industry 5.0 has made sustainability, human-centric and resilience the core of digital manufacturing, which also puts forward new requirements for the human-machine interaction (HMI) paradigm in human-centric smart manufacturing (HCSM). In the manufacturing scenario, the process of HMI can be divided into four parts: 1) Sensors and hardware, where the environment information and input signals are collected, 2) Data processing, where the signals are converted into data, 3) Transmission mechanism, where the data is transmitted to the processing centre, and 4) Interaction and collaboration. Among them, sensors and data are expected to become breakthrough points in optimising HMI. This is not only due to the emergence of new research, innovation and technologies but also because they are closely influenced by the new design concepts brought about by Industry 5.0. This paper analyses the latest studies and technologies in the sensor field and their possible applications in HCSM scenarios. Then, opportunities and challenges of data analysis in the HMI in Industry 5.0 are discussed. Finally, based on the design concepts and requirements of Industry 5.0, this paper demonstrates how they will become the key points for future HMI development. [ABSTRACT FROM AUTHOR]

Lack of service data for cardiac rehabilitation limits understanding of program delivery, benchmarking and quality improvement. This study aimed to describe current practices, management, utilisation and engagement with quality indicators in Australian programs. Cardiac rehabilitation programs (n=396) were identified from national directories and networks. Program coordinators were surveyed on service data capture, management systems and adoption of published national quality indicators. Text responses were coded and classified. Logistic regression determined independent associates of the use of data for quality improvement. A total of 319 (81%) coordinators completed the survey. Annual patient enrolments/programs were >200 (31.0%), 51−200 (46%) and ≤50 (23%). Most (79%) programs used an electronic system, alongside paper (63%) and/or another electronic system (19%), with 21% completely paper. While 84% knew of the national quality indicators, only 52% used them. Supplementary to patient care, data were used for reports to managers (57%) and funders (41%), to improve quality (56%), support funding (43%) and research (31%). Using data for quality improvement was more likely when enrolments where >200 (Odds ratio [OR] 3.83, 95% Confidence Interval [CI] 1.76−8.34) and less likely in Victoria (OR 0.24 95%, CI 0.08−0.77), New South Wales (OR 0.25 95%, CI 0.08−0.76) and Western Australia (OR 0.16 95%, CI 0.05−0.57). The collection of service data for cardiac rehabilitation patient data and its justification is diverse, limiting our capacity to benchmark and drive clinical practice. The findings strengthen the case for a national low-burden approach to data capture for quality care. [ABSTRACT FROM AUTHOR]

This paper outlines an example of Special Collections materials being used to support master's-level students' data and digital skills development. It evidences positive student learning outcomes through an unconventional interdepartmental collaboration between a digital media teaching team and Special Collections staff, which combined digital practices and tangible Special Collections material. This paper outlines the readings used for the course under review, the assignments used to evaluate students, student experiences of those assignments, and staff observations of student success and the logistics of course delivery. Findings show that in completing their assignments students (a) negotiated uncertainty; (b) increased understandings of data and digital methods; (c) embraced the lifeworlds of data; and (d) blurred the boundaries of analog and digital. More generally, this work finds that Special Collections material can be effectively used to support data and digital skills development, as well as help students gain broader understandings of digital contexts and media. Value is, however, dependent upon embracing uncertainty: something that universities tend to avoid. But, as is shown, staff and students alike can thrive in states of bewilderment. [ABSTRACT FROM AUTHOR]

This paper examines the effects of digital technologies on children's rights. All children's rights ought to be reinforced, respected, and protected. Majority of children devote their time on Internet using various social media platforms which have become a key means of teaching and learning. Increased internet access by children is associated with risks and escalation of children's rights violations. Institutions dealing with children continuously struggle to represent the rights of children enshrined within the Convention on the Rights of Children. The paper is mainly focuses on children's rights such as freedom of expression and information, privacy and data protection, protection and safety. Children's rights are not fully protected online in most cases particularly in countries like Botswana. The publication of sensitive information relating to children online is a violation of their rights to privacy and may affect their childhood development. The lack of relevant expert to investigate and prosecute cybercrime leads to the increase of cybercrime activities committed against children. [ABSTRACT FROM AUTHOR]

As a mesh-free approach, the incompressible Smoothed Particle Hydrodynamics (ISPH) method has been often used for simulating wave-structure interaction problems. In the conventional ISPH method, the pressure-projection phase of solving the pressure Poisson's equation (PPE) is the most time-consuming. In recent years, the machine learning (ML) techniques has gradually shown their potential in accelerating the computational fluid dynamics. In this paper, the graph neural network (GNN) supported ISPH method (ISPH_GNN), in which the GNN replaces solving the PPE for the fluid pressure in the conventional ISPH, is adopted for numerical simulations of wave-floater interactions. To the best of the authors' knowledge, this is the first work to study the wave-floater interactions by using GNN supported ISPH method. More importantly, this paper demonstrates that the GNN trained only on data for simpler wave-only cases (i.e. no structure in the wave fields) can be satisfactorily applied to the cases for wave-floater interactions. More specifically, the paper will show this by using the ISPH_GNN with such trained GNN model to simulate various different cases, including the decay tests of a box, a floating box subjected to a wave, the interaction between wave and a moored floating breakwater and the violent green water impact on a floating structure. In most of the cases, the numerical results are validated by comparing with experimental data. Agreement between them is surprisingly satisfactory, being as good as those obtained by the conventional ISPH. The paper will also show that the ISPH_GNN requires much less computational time (97 times less for the cases concerned) than the conventional ISPH for estimating pressure involved in wave-floater interactions. This reveals a great potential that one can train the GNN using the datasets for simpler cases and then use the ISPH_GNN to simulate wave-floater interaction problems. • The first work to study the wave-floater interactions by using the graph neural network (GNN) supported ISPH method. • The GNN trained only on data for simpler wave-only cases can be applied to more complex cases for wave-floater interactions. • The ISPH_GNN can give satisfactory results for the relatively complex cases involving wave-floater interactions. • The ISPH_GNN takes much less computation time than the conventional ISPH for estimating pressure. [ABSTRACT FROM AUTHOR]

Lameness is an important economic problem in the dairy sector, resulting in production loss and reduced welfare of dairy cows. Given the modern-day expansion of dairy herds, a tool to automatically detect lameness in real-time can therefore create added value for the farmer. The challenge in developing camera-based tools is that one system has to work for all the animals on the farm despite each animal having its own individual lameness response. Individualising these systems based on animal-level historical data is a way to achieve accurate monitoring on farm scale. The goal of this study is to optimise a lameness monitoring algorithm based on back posture values derived from a camera for individual cows by tuning the deviation thresholds and the quantity of the historical data being used. Back posture values from a sample of 209 Holstein Friesian cows in a large herd of over 2000 cows were collected during 15 months on a commercial dairy farm in Sweden. A historical data set of back posture values was generated for each cow to calculate an individual healthy reference per cow. For a gold standard reference, manual scoring of lameness based on the Sprecher scale was carried out weekly by a single skilled observer during the final 6 weeks of data collection. Using an individual threshold, deviations from the healthy reference were identified with a specificity of 82.3%, a sensitivity of 79%, an accuracy of 82%, and a precision of 36.1% when the length of the healthy reference window was not limited. When the length of the healthy reference window was varied between 30 and 250 days, it was observed that algorithm performance was maximised with a reference window of 200 days. This paper presents a high-performing lameness detection system and demonstrates the importance of the historical window length for healthy reference calculation in order to ensure the use of meaningful historical data in deviation detection algorithms. [ABSTRACT FROM AUTHOR]

Developing Urban Building Energy Models (UBEM) to manage heat decarbonisation at the neighbourhood level is both crucial and challenging. Our study addresses key research challenges: quantifying peak heat energy demand and developing reliable and replicable urban energy demand microsimulations. We applied our UBEM approach to a case study area of 228 houses served by the Ridgeway New Low Voltage (LV) electrical substation in Newcastle upon Tyne, to estimate the difference between existing peak heat and electricity demand at this scale. Results show that peak heat demand is 5 to 14 times greater than peak electricity demand, significantly exceeding estimates from national studies. Furthermore, our work follows a comprehensive validation framework, comparing our simulation results against all relevant and available UK datasets. This comparison demonstrates that our model is an overall good fit at the declared Level of Detail. However, our paper identifies significant challenges, particularly in validation, that need to be addressed to improve the reliability and usability of future UBEMs. Finally, we reflect on all our findings and make policy recommendations, as we believe addressing the issues raised in this paper is vital for enabling area-based planning of residential heating electrification, particularly in the UK. [ABSTRACT FROM AUTHOR]

Healthcare organizations have realized that Artificial intelligence (AI) can provide a competitive edge through personalized patient experiences, improved patient outcomes, early diagnosis, augmented clinician capabilities, enhanced operational efficiencies, or improved medical service accessibility. However, deploying AI-driven tools in the healthcare ecosystem could be challenging. This paper categorizes AI applications in healthcare and comprehensively examines the challenges associated with deploying AI in medical practices at scale. As AI continues to make strides in healthcare, its integration presents various challenges, including production timelines, trust generation, privacy concerns, algorithmic biases, and data scarcity. The paper highlights that flawed business models and wrong workflows in healthcare practices cannot be rectified merely by deploying AI-driven tools. Healthcare organizations should re-evaluate root problems such as misaligned financial incentives (e.g., fee-for-service models), dysfunctional medical workflows (e.g., high rates of patient readmissions), poor care coordination between different providers, fragmented electronic health records systems, and inadequate patient education and engagement models in tandem with AI adoption. This study also explores the need for a cultural shift in viewing AI not as a threat but as an enabler that can enhance healthcare delivery and create new employment opportunities while emphasizing the importance of addressing underlying operational issues. The necessity of investments beyond finance is discussed, emphasizing the importance of human capital, continuous learning, and a supportive environment for AI integration. The paper also highlights the crucial role of clear regulations in building trust, ensuring safety, and guiding the ethical use of AI, calling for coherent frameworks addressing transparency, model accuracy, data quality control, liability, and ethics. Furthermore, this paper underscores the importance of advancing AI literacy within academia to prepare future healthcare professionals for an AI-driven landscape. Through careful navigation and proactive measures addressing these challenges, the healthcare community can harness AI's transformative power responsibly and effectively, revolutionizing healthcare delivery and patient care. The paper concludes with a vision and strategic suggestions for the future of healthcare with AI, emphasizing thoughtful, responsible, and innovative engagement as the pathway to realizing its full potential to unlock immense benefits for healthcare organizations, physicians, nurses, and patients while proactively mitigating risks. • This study categorizes AI applications in healthcare and analyzes their deployment challenges at scale in medical practices. • The paper emphasizes a cultural shift to view AI as a healthcare delivery enhancer and job creator, not a threat. • To integrate AI, investments in finance and human capital, continuous learning, and a supportive environment are needed. • Clear regulatory frameworks should be developed to build trust, ensure safety, and guide the ethical use of AI in healthcare. • The study provides strategic suggestions for responsible AI use in healthcare to maximize benefits and minimize risks. [ABSTRACT FROM AUTHOR]

The way in which consumers engage with, utilise, or discard the technologies in their lives is constantly being reassessed and changed. This paper questions what role the emergent "right to repair" could play in resolving issues posed by the increasing ubiquity of the Internet of Things (IoT). The right gives consumers the ability and freedom to fix their devices, or to fair access to appropriate services that can carry out repair on their behalf. In this paper, firstly we establish the problem space surrounding consumer IoT – i.e., devices that are interconnected via the internet, enabling them to send and receive data. We reflect on hardware, software, and data components that pose legal and policy challenges for data protection, security, and sustainability. Through a literature review we then reflect on the current socio-legal developments that support or oppose changes in the consumer IoT market in regards to repair. We then highlight gaps in the existing literature that should inform future research trajectories in this area. This includes exploring disparities between environmental and consumer autonomy approaches, assessing consistency in regulatory developments, and market prioritisation. Finally, the paper concludes with a series of key insights and recommendations from our analysis including: recognition of the growing e-Waste problem and the inequalities it exacerbates and perpetuates; the need for identification and argumentation for different formulations of "repair" and how these may impact the implementation of a right going forward; the need for identification of the reasoning behind disparities in governmental approaches to the right to repair; and the need to practically translate better IoT design practices into reality. [ABSTRACT FROM AUTHOR]

Compared to traditional electronic maps, high-precision maps are more about providing services for vehicles. Because vehicles lack the visual recognition and logic inherent in humans, high-precision maps can assist vehicles in identifying obstacles, vehicles, etc., and can also expand their environmental awareness capabilities. This can provide vehicles with a global perspective that other sensors cannot obtain. This article includes road, traffic, and facility information outside the sensor monitoring range. This paper analyzes and defines road design data and high-precision map construction in detail, and then verifies the progressiveness of high-precision map construction methods based on road design data compared with traditional high-precision map construction methods through experimental analysis [ABSTRACT FROM AUTHOR]

Novel usages of brain stimulation combined with artificially intelligent (AI) systems promise to address a large range of diseases. These new conjoined technologies, such as brain-computer interfaces (BCI), are increasingly used in experimental and clinical settings to predict and alleviate symptoms of various neurological and psychiatric disorders. Due to their reliance on AI algorithms for feature extraction and classification, these BCI systems enable a novel, unprecedented, and direct connection between human cognition and artificial information processing. In this paper, we present the results of a study that investigates the phenomenology of human-machine symbiosis during a first-in-human experimental BCI trial designed to predict epileptic seizures. We employed qualitative semi-structured interviews to collect user experience data from a participant over a six-years period. We report on a clinical case where a specific embodied phenomenology emerged: namely, after BCI implantation, the patient reported experiences of increased agential capacity and continuity; and after device explantation, the patient reported persistent traumatic harms linked to agential discontinuity. To our knowledge, this is the first reported clinical case of a patient experiencing persistent agential discontinuity due to BCI explantation and potential evidence of an infringement on patient right, where the implanted person was robbed of her de novo agential capacities when the device was removed. • Some brain-computer interfaces predict seizures by monitoring neural data, some advise users based on specific neural data thresholds. • In the ethical literature, advisory devices are perceived as encompassing a stronger degree of autonomy because they keep users in the decisional loop. • We report on a clinical case where a BCI implantation yielded enhanced agential capacity and continuity, while device explantation caused ongoing traumatic harm and agential discontinuity • To our knowledge, the first reported case of a patient experiencing ongoing agential discontinuity due to BCI removal, raising concerns about potential patient rights violation. [ABSTRACT FROM AUTHOR]

The application of autonomous technology in food supply chains gives rise to a number of ethical considerations associated with the interaction between human and technology, human-technology-plant and human-technology-animal. These considerations and their implications influence technology design, the ways in which technology is applied, how the technology changes food supply chain practices, decision-making and the associated ethical aspects and outcomes. Using the concept of reflexive governance, this paper has critiqued existing reflective food-related ethical assessment tools and proposed the structural elements required for reflexive governance architectures which address both the sharing of data, and the use of artificial intelligence (AI) and machine learning in food supply chains. Considering the ethical implications of using autonomous technology in real life contexts is challenging. The current approach, focusing on discrete ethical elements in isolation e.g., ethical aspects or outcomes, normative standards or ethically orientated compliance-based business strategies, is not sufficient in itself. Alternatively, the application of more holistic, reflexive governance architectures can inform consideration of ethical aspects, potential ethical outcomes, in particular how they are interlinked and/or interdependent, and the need for mitigation at all lifecycle stages of technology and food product conceptualisation, design, realisation and adoption in the food supply chain. This research is of interest to those who are undertaking ethical deliberation on data sharing, and the use of AI and machine learning in food supply chains. • Autonomous technology can support decision-making in food supply chains. • The use of autonomous technology has ethical implications. • Ethical considerations focus on both aspects and outcomes of technology use. • Existing governance approaches are limited by being reflective, normative, rule-based. • Reflexive governance architectures for technology 'concept to realisation' are essential. [ABSTRACT FROM AUTHOR]

• The parameters can be identified successfully using incomplete frequency domain data. • The approach in this paper can result in good identification results from noisy data. • The optimal weight matrix indeed improves the identification accuracy. • The proposed approach is more efficient than that in the time domain-based approach. The technique to identify the system parameters thereof has attracted extensive research interest, since knowing the parameters would enable effective system control strategy and accurate response prediction. In this paper, a novel approach is developed to identify the parameters of the linear time-delay differential system by analyzing the complex system response in the frequency domain. Firstly, the complex frequency response of the time-delay system is expressed as a function of physical parameters and time-delay parameters, forming a typical optimization problem. Subsequently, the sensitivities with respect to the unknown parameters are derived. A novel sensitivity-based algorithm is adopted in the identification procedure. Trust-region constraint is implemented and hence tackled by Tikhonov regularization, which effectively enhances the efficiency of the algorithm. The feasibility and robustness of the identification procedure are evaluated by identifying the parameters of two numerical time-delay systems and an experimental case. [ABSTRACT FROM AUTHOR]

This paper provides significant contributions in the line of the so-called privacy-preserving OLAP research area, via extending the previous SPPOLAP's results provided recently. SPPOLAP is a state-of-the-art algorithm whose main goal consists in computing privacy-preserving OLAP data cubes effectively and efficiently. The main innovations carried-out by SPPOLAP are represented by the novel privacy OLAP notion and the flexible adoption of sampling-based techniques in order to achieve the final privacy-preserving data cube. In line with the main SPPOLAP's results, this paper significantly extends the previous research efforts by means of the following contributions: (i) complete algorithms of the whole SPPOLAP algorithmic framework; (ii) complexity analysis and results; (iii) comprehensive experimental analysis of SPPOLAP against real-life multidimensional data cubes, according to several experimental parameters. These contributions nice-fully complete the state-of-the-art SPPOLAP's results. [ABSTRACT FROM AUTHOR]

Previous studies show that time series data about the frequency of hits for tourism-related search terms from Google (Google Trends data) is a valuable predictor for short-term tourism demand forecasting in many different tourism regions worldwide. The paper contributes to this literature in three ways. First, it shows that Google Trends data is useful for short-term predictions of monthly tourist arrivals in several German holiday regions. Second, the paper also demonstrates that the Google Trends time series we employ share certain patterns with Google Trends time series used in previous studies, including several studies totally unrelated to the tourism industry. We refer to these artefacts as "spurious patterns" and perform a detailed analysis of their negative impact on forecasting. Last, the paper proposes a method to sanitize Google Trends data and reduce the adverse impact of spurious patterns, thereby paving the way to develop statistically sound tourism demand forecasts. • Google Trends data is a valuable predictor for tourism in Germany. • The data shows "spurious patterns" that are unrelated to tourism. • The spurious patterns have a negative impact on tourism forecasts. • Data modi_cation can reduce the impact of the spurious patterns. • The spurious patterns can be seen in several previously published studies. [ABSTRACT FROM AUTHOR]

• Summarize household energy consumption data characteristics and development trends. • Analyze household energy consumption patterns based on the time dimension. • Summarize the main influencing factors of household energy consumption. • Organically combine data, household energy consumption patterns and drivers. Understanding household energy consumption (HEC) demand patterns and their influencing factors at different times can provide insights into household energy demand and consumption behavior. Combining with optimization technology can improve energy efficiency, promote energy conservation, and predict future energy demand. Obtaining comprehensive and accurate data is the basis for revealing and analyzing HEC patterns and related activities. The main purpose of this paper is to review the current methods of obtaining HEC data in different time dimensions, HEC patterns, and influencing factors at different times, and organically combine all three. We first start with HEC data. This paper provides an overview of the current methods and data types used to collect information on HEC and related activities. It includes macro panel data, micro questionnaire survey data, micro case study data, and micro smart meter and advanced sensor data. Following this, presents current studies on HEC patterns, with a focus on cross-country comparisons in the temporal dimension. Direct and indirect (time-occupancy) studies are two ways of studying HEC patterns over time. Finally, discusses the factors that influence HEC demand, such as climate, building type, and socio-demographic factors. The understanding of these factors is crucial for promoting energy conservation and efficiency in households. [ABSTRACT FROM AUTHOR]

This paper takes the occasion of French DJ David Guetta's use of generative AI tools to create lyrics and a voice in the style of Eminem, which he then used in one of his concerts, as the basis for an exploration of the shifting meaning of creativity and originality in the age of generative AI. Our main contention is that the Guetta form of creativity with generative AI tools differs in certain important respects from what has come before. The paper describes an iterative, dynamic process of conception, prompting, generation, refining, and deployment to characterise creativity in this context. Nevertheless, we contend that copyright – specifically the concept of originality as articulated in US federal law – is a sufficiently durable legal mechanism that can manage these new cultural forms, and that the two basic requirements of modern copyright law (a tangible medium of expression and a modest degree of creativity) remain relevant in identifying the scope of legal protection. The paper argues that the David Guetta story reveals something more general about creativity in a digital age, namely that while hybrid-networked (i.e., human – corporate – machine) creators have always created hybrid-networked cultural forms (i.e., creations that blend human and technology-constituted elements), such hybridity becomes increasingly visible and complex in the context of a new world of generative AI. At the very least, earlier – and influential – models of creativity as human-driven involving creation ex nihilo become harder to sustain in a new age of generative AI. But this does not mean copyright or notions of originality are redundant or that copyright law cannot accommodate Guetta and other cases. Such an account seems important as it challenges the hegemonic and reductive view that AI "generates" artistic works autonomously and avoids reducing the copyright issues raised by such creative works to the related but distinct question of whether learning models rely on copyrighted data. As such, copyright law should remain an important mechanism to facilitate genuine creators who are using AI systems in innovative and unique ways to push the boundaries of their creativity. [ABSTRACT FROM AUTHOR]

Artificial intelligence is often framed as a tool to be used to counter violent extremism. This paper seeks to make a contribution to scholarship on the use of AI by considering the other side of the debate: how AI and algorithms themselves can and are being used to radicalize, polarize, and spread racism and political instability. The central argument of the paper is that AI and algorithms are not just tools deployed by national security agencies to prevent malicious activity online, but contributors to polarization, radicalism and political violence. Further to this, securitization processes have been instrumental in how AI has been designed, used and to the harmful otucomes that it has generated. The paper begins with an analysis of the connections between AI, polarization, radicalism and political violence. Drawing on the 'Copenhagen School' of International Relations theory, it then moves on to an empirical assessment of how AI has been securitized throughout its history, and in media and popular culture depictions, and by exploring a number of modern examples of AI having polarizing, radicalizing effects that have contributed to political violence. The third section of the article examines AI technology itself, arguing that problems exist in the design of AI, the data that it relies on, how it is used, and in its outcomes and impacts. The final section draws conclusions and policy implications, arguing that a reconceptualisation of AI-enabled security is necessary that is more attuned to the human, social and psychological impacts of the technology. • This paper analyses how AI and algorithms are being used to radicalize, polarize, and spread racism and political instability. • AI and algorithms are not just tools deployed by national security agencies but contributors to polarization, radicalism and political violence. • Securitization processes are a missing link to understanding how AI has been designed and used and to the harmful outcomes that it has generated. • Reconceptualizing AI-enabled conflict is necessary in a way that is more aware of the human, social and psychological impacts of the technology. [ABSTRACT FROM AUTHOR]

The energy of Wireless Sensor Network (WSN) in edge network is limited. In order to save energy, data bundling is proposed to reduce the energy consumed in setting up connection links by joining several packets into a single unit. However, although current data bundling schemes can reduce energy consumption, they will increase the delay of data routing, so it makes the multi-hop end to end delay a lot in WSN. Also, there is no appropriate solution to tackle the delayed growing problem caused by data bundling. And the energy consumption of the Acknowledgment (ACK) transmission is not comprehensively considered either. Regarding to those issues of previous researches, in this paper, a Data and ACK Bundling Scheme (DAB) is proposed to aggregate several data or ACK packets into a single unit to reduce the energy consumption as well as the end to end delay of the WSN. In the proposed DAB scheme, not only multiple packets but also ACK bundling into a single unit can reduce energy consumption. More importantly, the parameters of bundling technology are optimized from a global perspective in DAB scheme, so that WSN can not only improve the network lifetime, but also reduce the end to end delay which is impossible in previous studies. The theoretical analysis and experimental results illustrate that, while ensuring the reliability of data collection, our DAB scheme improves the energy efficiency by 29.03 % and reduces the delay by 22.29 % when compared with our traditional data collection scheme. [ABSTRACT FROM AUTHOR]

The manufacturing industry faces immense challenges for maintaining and increasing their productivity and flexibility. In this context, it is important for companies to ensure that their employees have the relevant data, information and knowledge necessary to make well-informed decisions. Due to recent development with Industry 4.0 enabling technologies that create new possibilities, the amount of available data, information and knowledge increase rapidly, but the insights into how to utilize it to its full potential are still lacking. In this paper, a human-centred perspective has been applied, aiming at improving how to cognitively support humans at work with new Industry 4.0 enabling technologies. Heavy emphasis is placed on people's requirements and preferences of data, information and knowledge for enhancing their performance and satisfaction at work. This paper examines the relationship between existing literature on dissemination of data, information and knowledge within the manufacturing industry with state-of-the-art research on Industry 4.0. The outcome of the research recognizes the increased importance of utilizing data, information and knowledge for people at work, facilitated by exploiting the new possibilities from Industry 4.0. To accomplish this, it is concluded that there exists an urgency to design: both a holistic framework for identifying and accommodating individuals' needs and expectations of relevant data, information and knowledge; and demonstrators and concepts to simplify the implementation of Industry 4.0 enabling technologies that support the aforementioned dissemination of data, information and knowledge. [ABSTRACT FROM AUTHOR]

This paper identifies the trends in the data used in pragmatics studies over the last 20 years, and thereby makes predictions about how pragmatics might develop in the next few years, other things being equal. To establish those trends, 200 papers from the Journal of Pragmatics, covering the period 1999 to 2018, were categorised. The categorisation scheme was designed to capture some of the key ways in which data in pragmatics varies, and included: the general focus of the paper (e.g. whether it is data-driven), the focal point in the data of the analysis, the quantity of the data, the medium of the data, the number of modes or channels represented in the data, the degree of interactivity of the data, the fictionality of the data, and the language of the data (specifically whether it only contains English). Trends discovered include: a continual strong focus on data, a shift of analytical focus towards more macro units, increasing use of greater quantities of data, decreasing dominance of purely spoken data, increasing use of multimodal data, increasing use of more interactive data, decreasing use of constructed examples, no increasing use of fictional data, and the continual diversification of the languages treated. • Pragmatics continues to be strongly linked to data, especially "real" data. • Pragmatics is shifting its analytical focus towards more macro, contextual units. • Pragmatics is engaging datasets of increasing size, though still avoids the very largest corpora. • Pragmatics is less dominated by purely spoken data, increasingly using multimodal data and digital media data, though the use of more interactive, dialogic data is increasing. • Pragmatics continues to diversify the range of languages it treats. [ABSTRACT FROM AUTHOR]

Abstract Among the sports fans beliefs about "hot hands" and "winning streaks" are widely spread, while the scientific debate about these effects is still ongoing. Recently in a paper by Ferreira (2018) detrended fluctuation analysis was applied to the NBA teams' win records. It was shown that 28 considered NBA teams exhibit persistence in the win record time series. In this paper we take the same data set and compare the obtained results against various random models. We find that the empirical results are consistent with the results obtained from various simple random models. Highlights • We apply DFA to study NBA regular season data. • For 26 of the 28 considered teams Hurst exponent is larger than 0.5, indicating persistence. • The obtained results are consistent with random models. • The observed persistence is simply an illusion caused by the limited amount of data. [ABSTRACT FROM AUTHOR]

Highlights • This paper presents a multi-channel biomedical data and information computing and visualization system that can efficiently process large histological image acquired from high-resolution microscopes. • Our system can dynamically display a volume of interest and extract tissue information using a new layer-based data navigation and exploration scheme. • During data exploring, the actual resolution of the data loaded can be dynamically determined and updated. The large data rendering is synchronized in four display windows at each data layer, where 2D textures are extracted from the Medical volume and mapped onto the displayed clipping planes with arbitrary orientations. • To test the efficiency and scalability of this system, we performed extensive evaluations using different hardware systems and large histological color datasets and experimental results demonstrated a solid performance. • Our system can deliver interactive data navigation speed and display detailed imaging information in real time, which is beyond the capability of commonly available biomedical data exploration software platforms. • Taking advantage of both CPU (central processing unit) main memory and GPU (graphics processing unit) graphics memory, this software platform can efficiently compute, process and visualize very large biomedical data and enhance data information. Abstract Background and objective Modern microscopes can acquire multi-channel large histological data from tissues of human beings or animals, which contain rich biomedical information for disease diagnosis and biological feature analysis. However, due to the large size, fuzzy tissue structure, and complicated multiple elements integrated in the image color space, it is still a challenge for current software systems to effectively calculate histological data, show the inner tissue structures and unveil hidden biomedical information. Therefore, we developed new algorithms and a software platform to address this issue. Methods This paper presents a multi-channel biomedical data computing and visualization system that can efficiently process large 3D histological images acquired from high-resolution microscopes. A novelty of our system is that it can dynamically display a volume of interest and extract tissue information using a layer-based data navigation scheme. During the data exploring process, the actual resolution of the loaded data can be dynamically determined and updated, and data rendering is synchronized in four display windows at each data layer, where 2D textures are extracted from the imaging volume and mapped onto the displayed clipping planes in 3D space. Results To test the efficiency and scalability of this system, we performed extensive evaluations using several different hardware systems and large histological color datasets acquired from a CryoViz 3D digital system. The experimental results demonstrated that our system can deliver interactive data navigation speed and display detailed imaging information in real time, which is beyond the capability of commonly available biomedical data exploration software platforms. Conclusion Taking advantage of both CPU (central processing unit) main memory and GPU (graphics processing unit) graphics memory, the presented software platform can efficiently compute, process and visualize very large biomedical data and enhance data information. The performance of this system can satisfactorily address the challenges of navigating and interrogating volumetric multi-spectral large histological image at multiple resolution levels. [ABSTRACT FROM AUTHOR]

Environmental horticulture firms supply products and services related to the design, installation, and maintenance of commercial and residential landscapes. An important issue in the environmental horticulture sector is the reduction and utilization of horticultural waste including plastic containers, trays, plastic greenhouse covers and field covers, and plastic bags. Using data collected through a survey in 2013, this paper identifies factors that influence the sector’s decision to participate in waste recycling programs in Georgia, USA. The study employs a decision model. Results indicate that significant determinants of the recycle decision include the firm size, motivation for recycling, current recycling situation, as well as owner/manager education level. To help firms to recycle, local governments, the environmental horticulture sector, and environmental protection organizations need to focus on lower-revenue and small size firms, broadly disseminate recycle program information, and enhance the local recycler’s services. [ABSTRACT FROM AUTHOR]

Highlights • Artificial Intelligence is already being used in some PR-related applications. • So far AI and PR have received little scholarly attention. • It is too early to form firm conclusions about AI impacts – but the public relations field needs to familiarize itself with developments now. Abstract Organizations of all kinds, as well as their in-house or agency public relations teams, increasingly co-opt Artificial Intelligence (AI) to enhance their capabilities. This paper examines a relatively new topic that has received little scholarly attention: the growing relationship between AI and public relations. It outlines several key roles that AI may play in future, based on trends in other industries, and considers the implications for public relations practitioners, their clients and employers. It therefore launches a dialogue about the diversity and extent of AI's uses in public relations practice. The paper argues that, to date, commentators have placed too much emphasis on AI's potential for task automation; AI's broader technological, economic and societal implications for public relations warrant greater critical attention. This does not imply that practitioners need become expert technologists; rather, they should develop a sufficient understanding of AI's present and potential uses to be able to offer informed counsel. [ABSTRACT FROM AUTHOR]

Automating traceability-based decision-making can shorten the reaction time to supply chain disruptions. This paper develops a framework for choosing automated decision-making (ADM) methods based on traceability data. It contains a toolbox comprising methods suitable for ADM, respective selection criteria and a new process to select a suitable ADM method based on companies' requirements. This process is based on an evaluation matrix matching methods and criteria. As a result, the ADM framework suggests the most suitable method to automate a specifically chosen decision. The developed framework is validated in the supply chain of a globally operating truck manufacturer. [ABSTRACT FROM AUTHOR]

Industry domains require distinct data and structures of building information models developed and tailored for their disciplines. To seamlessly exchange the building information models, Industry Foundation Classes (IFC), which is one of neutral formats, has been broadly used in the architecture, engineering and construction, and facility management industries. Model view definition (MVD), which is one of the IFC sub-schemas used by domain experts and BIM software vendors, consists of IFC-mapped data exchange requirements of each domain and helps software vendors develop IFC import and export features that allow project participants share and exchange BIM model information. Because of the heterogeneous translation processes and structures of IFC interfaces according to model views, their validation is imperative to ensure the integrity of BIM data and maintain a consistent data exchange environment. To accomplish this objective, this paper suggests the new approach to evaluating BIM data in accordance with diverse requirements of MVD. Since MVD entails various types of data exchange specifications, this research study examines their embedded checking rule types and categorizes corresponding implementation scenarios. In addition, this paper involves rule logic and IfcDoc-based BIM data validation developed based on the logical rule compositions of identified rules types and checking scenarios. This approach is expected to support sharing consistent BIM data sets and confirming the quality of received data pertaining to a syntax and semantics of a targeted model view. [ABSTRACT FROM AUTHOR]

Irrigation efficiency projects aim to conserve water for in-stream flow and agricultural use by reducing water losses throughout the system. Piping irrigation canals is a common irrigation efficiency method that results in trade-offs: while it increases efficiency of irrigation water conveyance, it reduces incidental groundwater recharge. This paper focuses on the data and decision-making of canal piping, focusing primarily on understanding the potential impacts of reduced canal leakage on shallow wells. By conducting a spatial analysis of shallow wells in the basin at risk of being impacted by canal piping, and combining this with interviews with water managers in central Oregon's Upper Deschutes Basin, we demonstrate the complex socio-natural dynamics and politics of water conservation decision-making. The research finds that irrigation canal piping is fully supported by water managers in the study area as a means of physically shifting flows of water towards particular valued uses, yet at the same time there is not enough data to understand the potential impacts of canal piping on water users reliant on canal seepage. Given the lack of localized shallow groundwater monitoring data, water managers are reliant upon basin-scale model predictions when defining the trade-offs in canal piping. Broadly, the research demonstrates that well-intended water management decisions can have trade-offs and impacts that are not well understood, pointing to the need for more groundwater monitoring and critical attention to the multiple scales of irrigation efficiency to inform the co-management of surface water and groundwater for the many water users within a basin. • Irrigation efficiency can impact basins in ways that are not understood. • Water managers rely on incomplete evidence when making efficiency decisions. • Irrigation efficiency efforts may negatively impact shallow domestic wells. • The politics of data gaps must be taken seriously in water management. • Critical lenses help to understand the complexity of irrigation efficiency. [ABSTRACT FROM AUTHOR]

The Analytic Hierarchy Process (AHP) is a de-facto standard technique in centralized decision making. Consider a situation where there is a need to rank a set of elements or alternatives, based on their value or utility, of which we just know pairwise relative information, i.e., the ratio of their values. AHP proved an effective tool to retrieve the value of each element, being able to handle also relative information affected by distortions, subjective biases and intransitivity. A downside of AHP, however, is that it requires complete information, i.e., knowledge on all pairs. In this paper, we extend the applicability of the AHP technique to the case of sparse information, i.e., when only a limited amount of information is available, and such an information corresponds to an undirected connected graph. We complement our sparse framework by developing novel criteria and metrics to evaluate the degree of consistency of the data at hand. Moreover, exploiting the proposed framework, we also provide a distributed formulation of AHP in which a set of agents, interacting through an undirected graph, are able to compute their own values (e.g., for ranking or leader election purposes), by only knowing the ratio of their values with respect to their neighbors. To this end, we develop a novel algorithm to let each agent i compute, the dominant eigenvalue and the i th component of the corresponding eigenvector of the sparse AHP matrix. We conclude the paper with a simulation campaign that numerically demonstrates the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Summary This paper explores the challenges for analysis of urbanization which can arise from insufficiently rigorous definition of what is “urban”. Policy makers and investors still use the ideas of “rural” versus “urban” and increasingly assume that the pace of urbanization in African countries is a measure of positive economic structural change. However most urban definitions do not incorporate economic characteristics. In Africa widely differing urban population thresholds and administrative factors are the most common criteria. The thresholds are often so low that many rural settlements are also defined as “urban” or they may be included on population density criteria, meaning the apparent pace of urbanization is inflated unrealistically. These issues are exemplified in this paper through detailed examples drawn from Kenya. It uses a range of sources including official census data and urban data published by Africapolis, as well as aerial images of rural and urban settlements in Kenya. It demonstrates how the use of population density criteria has inflated Kenyan urban data by the incorporation of very large numbers of rural people and explains how this can lead to entirely misleading interpretations of local and national urban and migration trends. Errors in urban figures can therefore have serious policy implications. It is argued that such errors can be reduced by not relying on a single criterion to define “urban” or by triangulating data on rural and urban settlements with other relevant information. [ABSTRACT FROM AUTHOR]

This paper focuses on an artificial immunity-enhancing module designed to counter internet-based cyberattacks on high-availability servers. The module consists of innate and adaptive immune functions. The innate immune function detects known and unknown cyberattacks, whereas the adaptive immune function uses a random forest classifier to learn the cyberattack detected by the innate immune function. This paper proposes a new innate immune function that detects two DoS attacks not detected by our previous innate immune function. In addition, a mechanism to maintain learning data is added to the adaptive immune function. The performance of the module was evaluated using four types of attack. Its overall detection accuracy was found to be 87.3%, corresponding to true positive and true negative rates of 78.95% and 95.70%, respectively. Investigation of its detection accuracy for four types of attack showed that a single type of attack degraded the overall detection accuracy. The overheads of the innate and adaptive immune functions were 6% and 4%, respectively, and were little affected by the number of trees in a random forest classifier. The number of learning data required by the adaptive immune function to maintain its high detection accuracy against cyberattacks was approximately 900. [ABSTRACT FROM AUTHOR]

Summary This paper analyzes the impact of data gap in Millennium Development Goals’ (MDGs) performance indicators on actual performance success of MDGs. Performance success, within the MDG framework, is quantified using six different ways proposed in the existing literature, including both absolute and relative performance and deviation from historical transition paths of MDG indicators. The empirical analysis clearly shows that the data gap in performance measurement is a significant predictor of poor MDG performance in terms of any of the six progress measures. Larger the data gap or weaker the performance measurement system, lesser is the probability of MDG performance success. The empirical methodology used in the paper combines a Heckman correction and instrumental variable estimation strategies to simultaneously account for potential endogeneity of the key data gap variable and bias due to sample selection. This result holds true even after controlling for overall national statistical capacity and a variety of socioeconomic factors. The paper underlines the need to strengthen the performance measurement system attached to the 2030 agenda for sustainable development and the associated Sustainable Development Goals (SDGs). This paper is the first attempt at empirically evaluating the value of data in the context of international development goals and gives empirical evidence for the need to harness the “data revolution” for sustainable development. [ABSTRACT FROM AUTHOR]

Over the last decades, researchers have studied one of the most complex combinatorial problems, i.e., the Concrete Delivery Problem (CDP), set different formulations of it, considered several objectives, and suggested various approaches. But no work has gathered this information to facilitate the comprehensive study of the CDP and investigate the research paths taken until today. Using a systematic search and mapping review, this paper fills that gap by (a) performing a comprehensive search on the CDP, and (b) mapping out and categorizing the existing literature. Specifically, this paper provides an overview of the various methods of the literature, categorizing the problem formulations based on the literature's different concepts, i.e., Integer Programming (IP), Time-Space Network, and the Job Shop model. The consistency between industry needs and existing constraints was discussed, and insights into the data sets corresponding to real-world cases were provided, identifying the necessary data for practitioners. [Display omitted] • A mapping review of methods for the Concrete Delivery Problem • Presentation of different formulations of the problem and explanation of the concept they are based on • Overview of the data sets and discussion on the data practitioners need for using any of the available optimization models [ABSTRACT FROM AUTHOR]

The deep sea has a large strategic depth, in which target detection is a crucial link, that is, how to detect the target submarine. For this reason, since this century, countries represented by the United States have innovated ideas, innovated underwater detection technology, and enhanced new underwater detection capabilities. In this paper, by consulting more than 100 literatures, the data acquisition technology in underwater acoustic detection in recent years has been investigated in detail, passive detection and active detection have been comprehensively reviewed, and the collaborative detection technology has also been summarized. In order to provide reference for related research. [ABSTRACT FROM AUTHOR]

Vibration data consists of batches of time series which if accumulated over a period of time is a huge collection of numeric data. Reducing such data for use in deep learning models for computational effciency is a challenge. Combinatorial and discrete approaches, on the other hand, is not an extensively explored area when it comes to datasets. This paper aims to identify feature reduction techniques based on discrete approaches such as euclidean distance using dot products on vibration data samples from accelerometers fitted on bearings. In this limited experimentation, the procured dataset by this approach is considerably smaller in size as compared to the actual complete data, and with comparable results in prediction models, it can be used as a smaller representation of a sensor timeline. The results based on different models show that such reductions can be considered in building IoT applications in industries based on sensors. [ABSTRACT FROM AUTHOR]

Resource allocation is difficult in large-scale systems in remote locations with sparse infrastructure and dynamic, demanding requirements, as can be the case in disaster management, emergency and oil spill response, and search and rescue operations. Resource allocation challenges are particularly acute in Arctic oil spill response, where the lack of infrastructure makes logistics challenging and heightens the need for comprehensive and thoughtful resource allocation models. This paper describes data challenges associated with the development of resource allocation models for dynamic network scheduling in support of Arctic oil spill response. A marine transportation traffic analysis and an Arctic resource allocation database, developed to support the resource allocation model development described, were developed using public and private information sources. This paper outlines the challenges associated with the data acquisition and analysis efforts, and provides recommendations for addressing the data challenges, which are similar to those in other complex, safety–critical resource allocation problem settings. [ABSTRACT FROM AUTHOR]

Inspired by the observation that a healthcare system usually involves various intelligent technologies from different disciplines, especially metaheuristics and data mining, this paper provides a brief survey of metaheuristics for healthcare system and a roadmap for researchers working on metaheuristics and healthcare to develop a more efficient and effective healthcare system. This paper begins with a discussion of changes for healthcare, followed by a brief review of the features of “up-to-date technologies for healthcare.”. Then, a learnable big data analytics framework for healthcare system is presented which provides a high performance solution to the forthcoming challenges of big data. Finally, changes, potentials, open issues, and future trends of metaheuristics for healthcare are addressed. [ABSTRACT FROM AUTHOR]

Patent landscape and the accompanying IP competitive intelligence involves understanding and anticipating the competitive environment within which a company operates. More specifically, IP competitive intelligence highlights emerging IP risks, provides patent portfolio benchmarking, monitors competitor technology development efforts, and predicts commercialization of technology. This paper provides a framework for patent landscape and IP competitive intelligence as driven by strategic intent. This paper advocates the benefits of both “quantitative” statistical analysis and “qualitative” human intelligence for IP competitive intelligence. Moreover, this paper defines four Levels of IP analysis with pruned examples for effective competitive intelligence. [ABSTRACT FROM AUTHOR]

The weight is assigned to all edges of graph according to a uniform policy and focus is on the leaves nodes of the Spanning Tree of Control Flow Graph of Program. If all leaves nodes of Spanning tree are covered by the test cases, then number of test cases required or testing the program can be reduced to great extent. Further the spanning tree is based on the concept of maximum spanning tree of CFG which helps in finding the critical paths of CFG. The chances of finding errors in critical are more as compare to the normal path. Covering of all the leaves nodes of Spanning Tree give guarantee for the coverage of Critical path also. Further the paper also shows that number of test cases required in case of Spanning tree is less as compare to the Dominance tree concept. For the generation of fast, unique and reliable test cases a optimization technique which is based on natural computing concept is used which is called as on Swine Influenza Models Based Optimization (SIMBO). The relation between nodes has been used as fitness function in this paper. Finally, the results in the paper show the effectiveness of SIMBO techniques as compare to Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). [ABSTRACT FROM AUTHOR]

Architects, engineers and contractors have been working with data for ages. What is new are the myriad ways they have to capture, analyze and apply the data that has been made readily available. Some practitioners and their organizations have balked at adding data to their repertoire, citing that they are unprepared to deal with data on top of everything else they have to contend with, including new technologies, work-processes and workflows. Data-centric organizations, on the other hand, have come to realize that capturing, engaging with, analyzing, and applying sustainability data in their projects is not “one more thing,” and have learned that to have the greatest impact, data cannot be added on top of an organization's on-going practices, but rather integrated into their pre-existing mind-sets and processes. This paper explores firms’ approaches and relationships to data, determining whether organizations are data-enabled, data-informed or data driven, or a hybrid, and how their approach can have an enormous influence on outcomes, especially where they pertain to sustainability. Likewise, many organizations remain document-centric, and are missing out on opportunities and greater gains that come with becoming data-centric. Innovative tools and insights to make the most of the data are already readily available. What has been missing, up until now, is an understanding of how others – academics and industry practitioners – are innovating behind the scenes to exploit these resources to great impact. Big data and predictive analytics are currently being leveraged across the globe to address wicked – complex and poorly defined – problems endemic of design and construction in the 21st century. Innovative results in leveraging data throughout the building lifecycle are a result of original research for a new book, “Data Driven Design and Construction: Strategies for Capturing, Analyzing and Applying Building Data,” (John Wiley & Sons, 2015.) Methods of Research: Because data driven design impacts so many facets of the building lifecycle, the research for this paper attempted to be as inclusive as possible. Using practice-based research and in-depth interviews with industry and academic leaders, the author relied on many sources, including first-hand interviews with 40 thought leaders in industry and education in the AECO industry who work day-to-day with data; with firm leaders and other industry executives at companies that range in size from sole proprietorships to large multinational organizations. The interviewee's responses were recorded, transcribed by the author, and condensed for publication. [ABSTRACT FROM AUTHOR]

Paired comparison data is often used to rank or order a set of items. In this paper we study a method for estimating the parameters associated with completely ordered cardinal paired comparison data. The analysis is carried out within the framework of graphical linear models but rather than using the least squares estimator, which may be difficult to analyze, we consider the average of all tree-based estimators for the connected comparison graph. The resulting estimator is a simple linear function of the sufficient statistics and has an easy to understand graph-theoretic interpretation. The statistical properties of this estimator are studied and it is shown to be unbiased, strongly consistent and asymptotically normal. Examples and numerical comparisons are provided and extensions are discussed. • Using a graphical linear model, we estimate parameters for paired comparison data. • We consider the average of all tree-based estimators in a connected comparison graph. • The resulting estimator is interpreted in terms of various graph-theoretic concepts. • The estimator is shown to be unbiased, strongly consistent and asymptotically normal. • Examples and simulation studies are provided along with some extensions. [ABSTRACT FROM AUTHOR]

In much of the literature on function approximation by deep networks, the function is assumed to be defined on some known domain, such as a cube or a sphere. In practice, the data might not be dense on these domains, and therefore, the approximation theory results are observed to be too conservative. In manifold learning, one assumes instead that the data is sampled from an unknown manifold; i.e., the manifold is defined by the data itself. Function approximation on this unknown manifold is then a two stage procedure: first, one approximates the Laplace–Beltrami operator (and its eigen-decomposition) on this manifold using a graph Laplacian, and next, approximates the target function using the eigen-functions. Alternatively, one estimates first some atlas on the manifold and then uses local approximation techniques based on the local coordinate charts. In this paper, we propose a more direct approach to function approximation on unknown , data defined manifolds without computing the eigen-decomposition of some operator or an atlas for the manifold, and without any kind of training in the classical sense. Our constructions are universal; i.e., do not require the knowledge of any prior on the target function other than continuity on the manifold. We estimate the degree of approximation. For smooth functions, the estimates do not suffer from the so-called saturation phenomenon. We demonstrate via a property called good propagation of errors how the results can be lifted for function approximation using deep networks where each channel evaluates a Gaussian network on a possibly unknown manifold. [ABSTRACT FROM AUTHOR]

We develop metrics for measuring the quality of synthetic health data for both education and research. We use novel and existing metrics to capture a synthetic dataset's resemblance, privacy, utility and footprint. Using these metrics, we develop an end-to-end workflow based on our generative adversarial network (GAN) method, HealthGAN, that creates privacy preserving synthetic health data. Our workflow meets privacy specifications of our data partner: (1) the HealthGAN is trained inside a secure environment; (2) the HealthGAN model is used outside of the secure environment by external users to generate synthetic data. This second step facilitates data handling for external users by avoiding de-identification, which may require special user training, be costly, or cause loss of data fidelity. This workflow is compared against five other baseline methods. While maintaining resemblance and utility comparable to other methods, HealthGAN provides the best privacy and footprint. We present two case studies in which our methodology was put to work in the classroom and research settings. We evaluate utility in the classroom through a data analysis challenge given to students and in research by replicating three different medical papers with synthetic data. Data, code, and the challenge that we organized for educational purposes are available. [ABSTRACT FROM AUTHOR]

Although person re-identification (ReID) has been intensively studied over the past few years, the shortage of annotated training data stands as an obstacle for further performance improvement. To address this issue, many data augmentation methods have been successfully applied to person ReID, such as random scaling, flipping and cropping, which mainly operate on a single image, whilst overlooking the relationship between images. Recently, generative adversarial networks (GANs) have been widely used for data augmentation and smoothly migrated to person ReID. However, the cost of training GAN is expensive and the performance improvement is often limited. Moreover, all these methods focus on augmenting samples for existing IDs. This paper proposes a simple yet effective data augmentation strategy based on self-supervised learning to handle these problems, which consists of the offline ID augmentation that can generate new categories and the online instance augmentation. These two components are integrated i nto a unified framework for boosting the ReID performance. Furthermore, a novel Siamese-like network is developed for ReID in conjunction with the proposed data augmentation method. Extensive experiments on three benchmark datasets demonstrate that the proposed approach outperforms the state-of-the-art by a clear margin, which verifies the robustness and the effectiveness of our method. Code will be released at: https://github.com/flychen321/data_aug_reid. [ABSTRACT FROM AUTHOR]