Florida Atlantic University

Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.

The Coding Manual fourth edition is reformatted to 15 chapters and divided into three parts, as opposed to six chapters in the third edition. This provides readers with more straightforward yet more detailed sectional references. The fourth edition also introduces two new first-cycle coding methods, 'metaphor coding' and 'theming the data categorically.' The fourth edition is aptly suited as a guide for teaching contexts for coding processes, given the provision of a companion website with student exercises and digital content. This includes screenshots of relevant academic software for analyzing the material and detailed thematic mapping diagrams. Furthermore, this book can assist novice researchers in determining the appropriate approaches to reinforce their perspectives and provide incremental processes for qualitative coding data.

The following paper presents current thinking and research on fit indices for structural equation modelling. The paper presents a selection of fit indices that are widely regarded as the most informative indices available to researchers. As well as outlining each of these indices, guidelines are presented on their use. The paper also provides reporting strategies of these indices and concludes with a discussion on the future of fit indices.

Machine learning and data mining techniques have been used in numerous real-world applications. An assumption of traditional machine learning methodologies is the training data and testing data are taken from the same domain, such that the input feature space and data distribution characteristics are the same. However, in some real-world machine learning scenarios, this assumption does not hold. There are cases where training data is expensive or difficult to collect. Therefore, there is a need to create high-performance learners trained with more easily obtained data from different domains. This methodology is referred to as transfer learning. This survey paper formally defines transfer learning, presents information on current solutions, and reviews applications applied to transfer learning. Lastly, there is information listed on software downloads for various transfer learning solutions and a discussion of possible future research work. The transfer learning solutions surveyed are independent of data size and can be applied to big data environments.

The authors integrate theory developed in several disciplines to determine five cognitive processes through which industrial buyers can develop trust of a supplier firm and its salesperson. These p...

In 2008 we published the first set of guidelines for standardizing research in autophagy. Since then, research on this topic has continued to accelerate, and many new scientists have entered the field. Our knowledge base and relevant new technologies have also been expanding. Accordingly, it is important to update these guidelines for monitoring autophagy in different organisms. Various reviews have described the range of assays that have been used for this purpose. Nevertheless, there continues to be confusion regarding acceptable methods to measure autophagy, especially in multicellular eukaryotes. For example, a key point that needs to be emphasized is thatthere is a difference between measurements that monitor the numbers or volume of autophagic elements (e.g., autophagosomes or autolysosomes) at any stage of the autophagic process versus those that measure flux through the autophagy pathway (i.e., the completeprocess including the amount and rate of cargo sequestered and degraded). In particular, a block in macroautophagy that results in autophagosome accumulation must be differentiated from stimuli that increase autophagic activity, defined as increasedautophagy induction coupled with increased delivery to, and degradation within, lysosomes (inmost higher eukaryotes and some protists such as Dictyostelium) or the vacuole (in plants and fungi). In other words, it is especially important that investigators new to the field understand that the appearance of more autophagosomes does not necessarily equate with more autophagy. In fact, in manycases, autophagosomes accumulate because of a block in trafficking to lysosomes without a concomitant change in autophagosome biogenesis, whereas an increase in autolysosomes may reflect a reduction in degradative activity. It is worth emphasizing here that lysosomal digestion is a stage of autophagy and evaluating its competence is a crucial part of the evaluation of autophagic flux, or complete autophagy. Here, we present a set of guidelines for the selection and interpretation of methods for use by investigators who aim to examine macroautophagy and related processes, as well as forreviewers who need to provide realistic and reasonable critiques of papers that are focused on these processes. These guidelines are not meant to be a formulaic set of rules, because the appropriate assays depend in part on the question being asked and the system being used. In addition, we emphasize that no individual assay is guaranteed to be the most appropriate one in every situation, and we strongly recommend the use of multipleassays to monitor autophagy. Along these lines, because of the potential for pleiotropic effects due to blocking autophagy through genetic manipulation, it is imperative to target by gene knockout or RNA interference more than one autophagyrelated protein. In addition, some individual Atg proteins, or groups of proteins, are involved in other cellular pathways implying that not all Atg proteins can be used as a specific marker for an autophagic process. In these guidelines, we consider these various methods of assessing autophagy and what information can, or cannot, be obtained from them. Finally, by discussing the merits and limits of particular assays, we hope to encourage technical innovation in the field.

We develop and test a theoretical model to investigate the assimilation of enterprise systems in the post-implementation stage within organizations. Specifically, this model explains how top management mediates the impact of external institutional pressures on the degree of usage of enterprise resource planning (ERP) systems. The hypotheses were tested using survey data from companies that have already implemented ERP systems. Results from partial least squares analyses suggest that mimetic pressures positively affect top management beliefs, which then positively affects top management participation in the ERP assimilation process. In turn, top management participation is confirmed to positively affect the degree of ERP usage. Results also suggest that coercive pressures positively affect top management participation without the mediation of top management beliefs. Surprisingly, we do not find support for our hypothesis that top management participation mediates the effect of normative pressures on ERP usage, but instead we find that normative pressures directly affect ERP usage. Our findings highlight the important role of top management in mediating the effect of institutional pressures on IT assimilation. We confirm that institutional pressures, which are known to be important for IT adoption and implementation, also contribute to post-implementation assimilation when the integration processes are prolonged and outcomes are dynamic and uncertain.

As one of the most successful approaches to building recommender systems, collaborative filtering ( CF ) uses the known preferences of a group of users to make recommendations or predictions of the unknown preferences for other users. In this paper, we first introduce CF tasks and their main challenges, such as data sparsity, scalability, synonymy, gray sheep, shilling attacks, privacy protection, etc., and their possible solutions. We then present three main categories of CF techniques: memory-based, model-based, and hybrid CF algorithms (that combine CF with other recommendation techniques), with examples for representative algorithms of each category, and analysis of their predictive performance and their ability to address the challenges. From basic techniques to the state-of-the-art, we attempt to present a comprehensive survey for CF techniques, which can be served as a roadmap for research and practice in this area.

The purpose of this study is to examine existing deep learning techniques for addressing class imbalanced data. Effective classification with imbalanced data is an important area of research, as high class imbalance is naturally inherent in many real-world applications, e.g., fraud detection and cancer detection. Moreover, highly imbalanced data poses added difficulty, as most learners will exhibit bias towards the majority class, and in extreme cases, may ignore the minority class altogether. Class imbalance has been studied thoroughly over the last two decades using traditional machine learning models, i.e. non-deep learning. Despite recent advances in deep learning, along with its increasing popularity, very little empirical work in the area of deep learning with class imbalance exists. Having achieved record-breaking performance results in several complex domains, investigating the use of deep neural networks for problems containing high levels of class imbalance is of great interest. Available studies regarding class imbalance and deep learning are surveyed in order to better understand the efficacy of deep learning when applied to class imbalanced data. This survey discusses the implementation details and experimental results for each study, and offers additional insight into their strengths and weaknesses. Several areas of focus include: data complexity, architectures tested, performance interpretation, ease of use, big data application, and generalization to other domains. We have found that research in this area is very limited, that most existing work focuses on computer vision tasks with convolutional neural networks, and that the effects of big data are rarely considered. Several traditional methods for class imbalance, e.g. data sampling and cost-sensitive learning, prove to be applicable in deep learning, while more advanced methods that exploit neural network feature learning abilities show promising results. The survey concludes with a discussion that highlights various gaps in deep learning from class imbalanced data for the purpose of guiding future research.

autophagic responses. Here, we critically discuss current methods of assessing autophagy and the information they can, or cannot, provide. Our ultimate goal is to encourage intellectual and technical innovation in the field.

Despite a decade since the inception of B2C e-commerce, the uncertainty of the online environment still makes many consumers reluctant to engage in online exchange relationships. Even if uncertainty has been widely touted as the primary barrier to online transactions, the literature has viewed uncertainty as a “background” mediator with insufficient conceptualization and measurement. To better understand the nature of uncertainty and mitigate its potentially harmful effects on B2C e-commerce adoption (especially for important purchases), this study draws upon and extends the principal-agent perspective to identify and propose a set of four antecedents of perceived uncertainty in online buyer–seller relationships—perceived information asymmetry, fears of seller opportunism, information privacy concerns, and information security concerns—which are drawn from the agency problems of adverse selection (hidden information) and moral hazard (hidden action). To mitigate uncertainty in online exchange relationships, this study builds upon the principal–agent perspective to propose a set of four uncertainty mitigating factors—trust, website informativeness, product diagnosticity, and social presence— that facilitate online exchange relationships by overcoming the agency problems of hidden information and hidden action through the logic of signals and incentives. The proposed structural model is empirically tested with longitudinal data from 521 consumers for two products (prescription drugs and books) that differ on their level of purchase involvement. The results support our model, delineating the process by which buyers engage in online exchange relationships by mitigating uncertainty. Interestingly, the proposed model is validated for two distinct targets, a specific website and a class of websites. Implications for understanding and facilitating online exchange relationships for different types of purchases, mitigating uncertainty perceptions, and extending the principal– agent perspective are discussed.

Abstract Big Data Analytics and Deep Learning are two high-focus of data science. Big Data has become important as many organizations both public and private have been collecting massive amounts of domain-specific information, which can contain useful information about problems such as national intelligence, cyber security, fraud detection, marketing, and medical informatics. Companies such as Google and Microsoft are analyzing large volumes of data for business analysis and decisions, impacting existing and future technology. Deep Learning algorithms extract high-level, complex abstractions as data representations through a hierarchical learning process. Complex abstractions are learnt at a given level based on relatively simpler abstractions formulated in the preceding level in the hierarchy. A key benefit of Deep Learning is the analysis and learning of massive amounts of unsupervised data, making it a valuable tool for Big Data Analytics where raw data is largely unlabeled and un-categorized. In the present study, we explore how Deep Learning can be utilized for addressing some important problems in Big Data Analytics, including extracting complex patterns from massive volumes of data, semantic indexing, data tagging, fast information retrieval, and simplifying discriminative tasks. We also investigate some aspects of Deep Learning research that need further exploration to incorporate specific challenges introduced by Big Data Analytics, including streaming data, high-dimensional data, scalability of models, and distributed computing. We conclude by presenting insights into relevant future works by posing some questions, including defining data sampling criteria, domain adaptation modeling, defining criteria for obtaining useful data abstractions, improving semantic indexing, semi-supervised learning, and active learning.

While privacy is a highly cherished value, few would argue with the notion that absolute privacy is unattainable. Individuals make choices in which they surrender a certain degree of privacy in exchange for outcomes that are perceived to be worth the risk of information disclosure. This research attempts to better understand the delicate balance between privacy risk beliefs and confidence and enticement beliefs that influence the intention to provide personal information necessary to conduct transactions on the Internet. A theoretical model that incorporated contrary factors representing elements of a privacy calculus was tested using data gathered from 369 respondents. Structural equations modeling (SEM) using LISREL validated the instrument and the proposed model. The results suggest that although Internet privacy concerns inhibit e-commerce transactions, the cumulative influence of Internet trust and personal Internet interest are important factors that can outweigh privacy risk perceptions in the decision to disclose personal information when an individual uses the Internet. These findings provide empirical support for an extended privacy calculus model.

Big Data concern large-volume, complex, growing data sets with multiple, autonomous sources. With the fast development of networking, data storage, and the data collection capacity, Big Data are now rapidly expanding in all science and engineering domains, including physical, biological and biomedical sciences. This paper presents a HACE theorem that characterizes the features of the Big Data revolution, and proposes a Big Data processing model, from the data mining perspective. This data-driven model involves demand-driven aggregation of information sources, mining and analysis, user interest modeling, and security and privacy considerations. We analyze the challenging issues in the data-driven model and also in the Big Data revolution.

The hypothesis was tested that children whose families differ in socioeconomic status (SES) differ in their rates of productive vocabulary development because they have different language-learning experiences. Naturalistic interaction between 33 high-SES and 30 mid-SES mothers and their 2-year-old children was recorded at 2 time points 10 weeks apart. Transcripts of these interactions provided the basis for estimating the growth in children's productive vocabularies between the first and second visits and properties of maternal speech at the first visit. The high-SES children grew more than the mid-SES children in the size of their productive vocabularies. Properties of maternal speech that differed as a function of SES fully accounted for this difference. Implications of these findings for mechanisms of environmental influence on child development are discussed.

Empirical studies and some high-profile anecdotal cases have demonstrated a link between suicidal ideation and experiences with bullying victimization or offending. The current study examines the extent to which a nontraditional form of peer aggression--cyberbullying--is also related to suicidal ideation among adolescents. In 2007, a random sample of 1,963 middle-schoolers from one of the largest school districts in the United States completed a survey of Internet use and experiences. Youth who experienced traditional bullying or cyberbullying, as either an offender or a victim, had more suicidal thoughts and were more likely to attempt suicide than those who had not experienced such forms of peer aggression. Also, victimization was more strongly related to suicidal thoughts and behaviors than offending. The findings provide further evidence that adolescent peer aggression must be taken seriously both at school and at home, and suggest that a suicide prevention and intervention component is essential within comprehensive bullying response programs implemented in schools.

To date, many important threads of information privacy research have developed, but these threads have not been woven together into a cohesive fabric. This paper provides an interdisciplinary review of privacy-related research in order to enable a more cohesive treatment. With a sample of 320 privacy articles and 128 books and book sections, we classify previous literature in two ways: (1) using an ethics-based nomenclature of normative, purely descriptive, and empirically descriptive, and (2) based on their level of analysis: individual, group, organizational, and societal. Based upon our analyses via these two classification approaches, we identify three major areas in which previous research contributions reside: the conceptualization of information privacy, the relationship between information privacy and other constructs, and the contextual nature of these relationships. As we consider these major areas, we draw three overarching conclusions. First, there are many theoretical developments in the body of normative and purely descriptive studies that have not been addressed in empirical research on privacy. Rigorous studies that either trace processes associated with, or test implied assertions from, these value-laden arguments could add great value. Second, some of the levels of analysis have received less attention in certain contexts than have others in the research to date. Future empirical studies—both positivist and interpretive—could profitably be targeted to these under-researched levels of analysis. Third, positivist empirical studies will add the greatest value if they focus on antecedents to privacy concerns and on actual outcomes. In that light, we recommend that researchers be alert to an overarching macro model that we term APCO (Antecedents → Privacy Concerns → Outcomes).

BACKGROUND: Randomized trials have shown that low-dose aspirin decreases the risk of a first myocardial infarction in men, with little effect on the risk of ischemic stroke. There are few similar data in women. METHODS: We randomly assigned 39,876 initially healthy women 45 years of age or older to receive 100 mg of aspirin on alternate days or placebo and then monitored them for 10 years for a first major cardiovascular event (i.e., nonfatal myocardial infarction, nonfatal stroke, or death from cardiovascular causes). RESULTS: During follow-up, 477 major cardiovascular events were confirmed in the aspirin group, as compared with 522 in the placebo group, for a nonsignificant reduction in risk with aspirin of 9 percent (relative risk, 0.91; 95 percent confidence interval, 0.80 to 1.03; P=0.13). With regard to individual end points, there was a 17 percent reduction in the risk of stroke in the aspirin group, as compared with the placebo group (relative risk, 0.83; 95 percent confidence interval, 0.69 to 0.99; P=0.04), owing to a 24 percent reduction in the risk of ischemic stroke (relative risk, 0.76; 95 percent confidence interval, 0.63 to 0.93; P=0.009) and a nonsignificant increase in the risk of hemorrhagic stroke (relative risk, 1.24; 95 percent confidence interval, 0.82 to 1.87; P=0.31). As compared with placebo, aspirin had no significant effect on the risk of fatal or nonfatal myocardial infarction (relative risk, 1.02; 95 percent confidence interval, 0.84 to 1.25; P=0.83) or death from cardiovascular causes (relative risk, 0.95; 95 percent confidence interval, 0.74 to 1.22; P=0.68). Gastrointestinal bleeding requiring transfusion was more frequent in the aspirin group than in the placebo group (relative risk, 1.40; 95 percent confidence interval, 1.07 to 1.83; P=0.02). Subgroup analyses showed that aspirin significantly reduced the risk of major cardiovascular events, ischemic stroke, and myocardial infarction among women 65 years of age or older. CONCLUSIONS: In this large, primary-prevention trial among women, aspirin lowered the risk of stroke without affecting the risk of myocardial infarction or death from cardiovascular causes, leading to a nonsignificant finding with respect to the primary end point.

Class imbalance is a problem that is common to many application domains. When examples of one class in a training data set vastly outnumber examples of the other class(es), traditional data mining algorithms tend to create suboptimal classification models. Several techniques have been used to alleviate the problem of class imbalance, including data sampling and boosting. In this paper, we present a new hybrid sampling/boosting algorithm, called RUSBoost, for learning from skewed training data. This algorithm provides a simpler and faster alternative to SMOTEBoost, which is another algorithm that combines boosting and data sampling. This paper evaluates the performances of RUSBoost and SMOTEBoost, as well as their individual components (random undersampling, synthetic minority oversampling technique, and AdaBoost). We conduct experiments using 15 data sets from various application domains, four base learners, and four evaluation metrics. RUSBoost and SMOTEBoost both outperform the other procedures, and RUSBoost performs comparably to (and often better than) SMOTEBoost while being a simpler and faster technique. Given these experimental results, we highly recommend RUSBoost as an attractive alternative for improving the classification performance of learners built using imbalanced data.

BACKGROUND: Coronary heart disease (CHD) is the most common cause of death globally. However, with falling CHD mortality rates, an increasing number of people living with CHD may need support to manage their symptoms and prognosis. Exercise-based cardiac rehabilitation (CR) aims to improve the health and outcomes of people with CHD. This is an update of a Cochrane Review previously published in 2016. OBJECTIVES: To assess the clinical effectiveness and cost-effectiveness of exercise-based CR (exercise training alone or in combination with psychosocial or educational interventions) compared with 'no exercise' control, on mortality, morbidity and health-related quality of life (HRQoL) in people with CHD. SEARCH METHODS: We updated searches from the previous Cochrane Review, by searching CENTRAL, MEDLINE, Embase, and two other databases in September 2020. We also searched two clinical trials registers in June 2021. SELECTION CRITERIA: We included randomised controlled trials (RCTs) of exercise-based interventions with at least six months' follow-up, compared with 'no exercise' control. The study population comprised adult men and women who have had a myocardial infarction (MI), coronary artery bypass graft (CABG) or percutaneous coronary intervention (PCI), or have angina pectoris, or coronary artery disease. DATA COLLECTION AND ANALYSIS: = 53%), and of small study bias for all-cause hospitalisation, but not for all other outcomes. At medium-term follow-up, although there may be little to no difference in all-cause mortality (RR 0.90, 95% CI 0.80 to 1.02; 15 trials), MI (RR 1.07, 95% CI 0.91 to 1.27; 12 trials), PCI (RR 0.96, 95% CI 0.69 to 1.35; 6 trials), CABG (RR 0.97, 95% CI 0.77 to 1.23; 9 trials), and all-cause hospitalisation (RR 0.92, 95% CI 0.82 to 1.03; 9 trials), a large reduction in cardiovascular mortality was found (RR 0.77, 95% CI 0.63 to 0.93; 5 trials). Evidence is uncertain for difference in risk of cardiovascular hospitalisation (RR 0.92, 95% CI 0.76 to 1.12; 3 trials). At long-term follow-up, although there may be little to no difference in all-cause mortality (RR 0.91, 95% CI 0.75 to 1.10), exercise-based CR may result in a large reduction in cardiovascular mortality (RR 0.58, 95% CI 0.43 to 0.78; 8 trials) and MI (RR 0.67, 95% CI 0.50 to 0.90; 10 trials). Evidence is uncertain for CABG (RR 0.66, 95% CI 0.34 to 1.27; 4 trials), and PCI (RR 0.76, 95% CI 0.48 to 1.20; 3 trials). Meta-regression showed benefits in outcomes were independent of CHD case mix, type of CR, exercise dose, follow-up length, publication year, CR setting, study location, sample size or risk of bias. There was evidence that exercise-based CR may slightly increase HRQoL across several subscales (SF-36 mental component, physical functioning, physical performance, general health, vitality, social functioning and mental health scores) up to 12 months' follow-up; however, these may not be clinically important differences. The eight trial-based economic evaluation studies showed exercise-based CR to be a potentially cost-effective use of resources in terms of gain in quality-adjusted life years (QALYs). AUTHORS' CONCLUSIONS: This updated Cochrane Review supports the conclusions of the previous version, that exercise-based CR provides important benefits to people with CHD, including reduced risk of MI, a likely small reduction in all-cause mortality, and a large reduction in all-cause hospitalisation, along with associated healthcare costs, and improved HRQoL up to 12 months' follow-up. Over longer-term follow-up, benefits may include reductions in cardiovascular mortality and MI. In the last decade, trials were more likely to include females, and be undertaken in LMICs, increasing the generalisability of findings. Well-designed, adequately-reported RCTs of CR in people with CHD more representative of usual clinical practice are still needed. Trials should explicitly report clinical outcomes, including mortality and hospital admissions, and include validated HRQoL outcome measures, especially over longer-term follow-up, and assess costs and cost-effectiveness.