Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Machine learning has made significant advances in the field of image processing. The foundation of this success is supervised learning, which necessitates annotated labels generated by humans and hence learns from labelled data, whereas unsupervised learning learns from unlabeled data. Self-supervised learning (SSL) is a type of un-supervised learning that helps in the performance of downstream computer vision tasks such as object detection, image comprehension, image segmentation, and so on. It can develop generic artificial intelligence systems at a low cost using unstructured and unlabeled data. The authors of this review article have presented detailed literature on self-supervised learning as well as its applications in different domains. The primary goal of this review article is to demonstrate how images learn from their visual features using self-supervised approaches. The authors have also discussed various terms used in self-supervised learning as well as different types of learning, such as contrastive learning, transfer learning, and so on. This review article describes in detail the pipeline of self-supervised learning, including its two main phases: pretext and downstream tasks. The authors have shed light on various challenges encountered while working on self-supervised learning at the end of the article. PubDate: 2023-01-20
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Disordered solids are ubiquitous in engineering and everyday use. Although research has made considerable progress in the last decades, our understanding of the mechanics of these materials is, at best, in an embryonic state. Since the nature of disorder complicates the realization of physically meaningful continuum-mechanical models, particle-based molecular descriptions provide a powerful alternative. This paper reviews the numerical realization of classical molecular dynamics from an engineer’s perspective, starting with selecting potential functions, boundary conditions, time integration, and thermodynamic ensembles. Then, we discuss the concept of the potential energy landscape and the computational realization of the most suitable minimization methods. Subsequently, we discuss the algorithms necessary to numerically generate disordered materials, considering their thermodynamic properties and structural identification. We comprehensively and critically review computational methods and strategies available to mimic disordered materials on a molecular level and discuss some intriguing phenomena that are, to date, mostly ignored when applying models based on continuum-mechanical frameworks. We present the crucial difference between the shear response of a crystalline and a disordered structure. In this context, we elaborate on why it is beneficial to use an overdamped, athermal description to disentangle the complex deformation mechanics of disordered solids and comprehensively discuss the theory of the mechanics of disordered materials, including the problems of prediction and reversibility. Furthermore, we examine the fracture process on the nanoscale and investigate the response behavior to more complex deformation protocols. Finally, we provide critical conclusions, including challenges and future perspectives for engineers. PubDate: 2023-01-20
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract In this paper we will present a review of recent advances in the application of the augmented Lagrange multiplier method as a general approach for generating multiplier-free stabilised methods. The augmented Lagrangian method consists of a standard Lagrange multiplier method augmented by a penalty term, penalising the constraint equations, and is well known as the basis for iterative algorithms for constrained optimisation problems. Its use as a stabilisation methods in computational mechanics has, however, only recently been appreciated. We first show how the method generates Galerkin/Least Squares type schemes for equality constraints and then how it can be extended to develop new stabilised methods for inequality constraints. Application to several different problems in computational mechanics is given. PubDate: 2023-01-20
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Machine learning is an established data interpretation tool for the development of processing, extracting and extrapolating evocative results from complex data sets. Data-based and computer-aided cancer research in patients expands at a rapid pace and presents a growing landscape of potential for Machine Learning methodologies, driven by the growing need for personalization of medical procedures. Previous research on artificial neural networks displays remarkable improvements in data mining tools and superior computational performance in prediction and diagnostics of cancer. For data mining, the article initially reviews machine-learning tools available for detection, susceptibility, reoccurrence and prediction of cancer prognosis. This article summarizes major challenges and problem-solving methods with examples of tools and brief description of algorithms used to improve the efficiency of cancer treatment and the development of personalized medicine and treatment for diverse types of cancer-based on genomic and protein data. PubDate: 2023-01-20
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The rapid growth of IoT devices leads to increasing requests. These tremendous requests cannot be processed by IoT devices due to the computational power of IoT devices and the disparate requirements of requests. Cloud computing seemed appealing to service these requests due to its remarkable characteristics. However, the physical gap between the Cloud datacenter and IoT devices causes a huge latency overhead. Furthermore, the centralized datacenter also experiences tremendous power consumption. Therefore, the Fog computing layer is introduced as a complementary layer to Cloud computing in between the IoT and the Cloud layer. Fog computing appears as cutting-edge technology to leverage the large computations in the Fog layer, thereby minimizing the latency gap and the power consumption of the datacenters. A Mist layer is placed in between the Fog and IoT layer to enable routing of the requests to Fog nodes and Cloud virtual machines. Many articles propose different load balancing strategies to distribute the loads uniformly in both Fog and Cloud layers. This contribution considers a wide spectrum of reviews as well as research articles into consideration ranging from 2010 to 2022. Besides, a layered architecture is proposed considering the IoT, Mist, Fog, and Cloud layers. Furthermore, research queries are analyzed and answered about the load balancing for these evolving paradigms, critical issues and challenges, and future directions. It is believed that this contribution would be a helping hand for the nascent researchers to get an insight into evolving paradigms, algorithms, issues, challenges, and future directions. PubDate: 2023-01-19
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Image-Based Simulation (IBSim) is the process by which a digital representation of a real geometry is generated from image data for the purpose of performing a simulation with greater accuracy than with idealised Computer Aided Design (CAD) based simulations. Whilst IBSim originates in the biomedical field, the wider adoption of imaging for non-destructive testing and evaluation (NDT/NDE) within the High-Value Manufacturing (HVM) sector has allowed wider use of IBSim in recent years. IBSim is invaluable in scenarios where there exists a non-negligible variation between the ‘as designed’ and ‘as manufactured’ state of parts. It has also been used for characterisation of geometries too complex to accurately draw with CAD. IBSim simulations are unique to the geometry being imaged, therefore it is possible to perform part-specific virtual testing within batches of manufactured parts. This novel review presents the applications of IBSim within HVM, whereby HVM is the value provided by a manufactured part (or conversely the potential cost should the part fail) rather than the actual cost of manufacturing the part itself. Examples include fibre and aggregate composite materials, additive manufacturing, foams, and interface bonding such as welding. This review is divided into the following sections: Material Characterisation; Characterisation of Manufacturing Techniques; Impact of Deviations from Idealised Design Geometry on Product Design and Performance; Customisation and Personalisation of Products; IBSim in Biomimicry. Finally, conclusions are drawn, and observations made on future trends based on the current state of the literature. PubDate: 2023-01-18
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The absolute previously infected novel coronavirus (COVID-19) was found in Wuhan, China, in December 2019. The COVID-19 epidemic has spread to more than 220 nations and territories globally and has altogether influenced each part of our day-to-day lives. As of 9th March 2022, a total aggregate of 44,78,82,185 (60,07,317) contaminated (dead) COVID-19 cases were accounted for all over the world. The quantities of contaminated cases passing despite everything increment essentially and do not indicate a controlled circumstance. The scope of this paper is to address this issue by presenting a comprehensive and comparative analysis of the existing Machine Learning (ML), Deep Learning (DL) and Artificial Intelligence (AI) based approaches used in significance in reacting to the COVID-19 epidemic and diagnosing the severe impacts. The paper provides, firstly, an overview of COVID-19 infection and highlights of this article; Secondly, an overview of exploring various executive innovations by utilizing different resources to stop the spread of COVID-19; Thirdly, a comparison of existing predicting methods of COVID-19 in the literature, with focus on ML, DL and AI-driven techniques with performance metrics; and finally, a discussion on the results of the work as well as future scope. PubDate: 2023-01-17
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract With Artificial Intelligence advancement, Deep neural networks (DNN) are extensively used for decision-making in intelligent systems. However, improved performance and accuracy have been achieved through the increasing use of complex models, which makes it challenging for users to understand and trust. This ambiguous nature of these Deep machine learning models of high accuracy and low interpretability is problematic for their adoption in critical domains where it is vital to be able to explain the decisions made by the system. Explainable Artificial Intelligence has become an exciting field for explaining and interpreting machine learning models. Among the different data types used in machine learning, image data is considered hard to train because of the factors such as class, scale, viewpoint, and background variations. This paper aims to provide a rounded view of emerging methods to explain DNN models as a way to boost transparency in image-based deep learning with the analysis of the current and upcoming trends. PubDate: 2023-01-13
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Meta-heuristic algorithms have a high position among academic researchers in various fields, such as science and engineering, in solving optimization problems. These algorithms can provide the most optimal solutions for optimization problems. This paper investigates a new meta-heuristic algorithm called Slime Mould algorithm (SMA) from different optimization aspects. The SMA algorithm was invented due to the fluctuating behavior of slime mold in nature. It has several new features with a unique mathematical model that uses adaptive weights to simulate the biological wave. It provides an optimal pathway for connecting food with high exploration and exploitation ability. As of 2020, many types of research based on SMA have been published in various scientific databases, including IEEE, Elsevier, Springer, Wiley, Tandfonline, MDPI, etc. In this paper, based on SMA, four areas of hybridization, progress, changes, and optimization are covered. The rate of using SMA in the mentioned areas is 15, 36, 7, and 42%, respectively. According to the findings, it can be claimed that SMA has been repeatedly used in solving optimization problems. As a result, it is anticipated that this paper will be beneficial for engineers, professionals, and academic scientists. PubDate: 2023-01-12
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Software development analytics is a research area concerned with providing insights to improve product deliveries and processes. Many types of studies, data sources and mining methods have been used for that purpose. This systematic literature review aims at providing an aggregate view of the relevant studies on Software Development Analytics in the past decade, with an emphasis on its application in practical settings. Definition and execution of a search string upon several digital libraries, followed by a quality assessment criteria to identify the most relevant papers. On those, we extracted a set of characteristics (study type, data source, study perspective, development life-cycle activities covered, stakeholders, mining methods, and analytics scope) and classified their impact against a taxonomy. Source code repositories, exploratory case studies, and developers are the most common data sources, study types, and stakeholders, respectively. Testers also get moderate attention from researchers. Product managers’ concerns are being addressed frequently and project managers are also present but with less prevalence. Mining methods are rapidly evolving, as reflected in their identified long list. Descriptive statistics are the most usual method followed by correlation analysis. Being software development an important process in every organization, it was unexpected to find that process mining was present in only one study. Most contributions to the software development life cycle were given in the quality dimension. Time management and costs control were less prevalent. The analysis of security aspects is even more reduced, however, evidences suggest it is an increasing topic of concern. Risk management contributions are also scarce. There is a wide improvement margin for software development analytics in practice. For instance, mining and analyzing the activities performed by software developers in their actual workbench, i.e., in their IDEs. Together with mining developers’ behaviors, based on the evidences and trend, in a short term period we expect an increase in the volume of studies related with security and risks management. PubDate: 2023-01-10
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Hyperelasticity is a common modeling approach to reproduce the nonlinear mechanical behavior of rubber materials at finite deformations. It is not only employed for stand-alone, purely elastic models but also within more sophisticated frameworks like viscoelasticity or Mullins-type softening. The choice of an appropriate strain energy function and identification of its parameters is of particular importance for reliable simulations of rubber products. The present manuscript provides an overview of suitable hyperelastic models to reproduce the isochoric as well as volumetric behavior of nine widely used rubber compounds. This necessitates firstly a discussion on the careful preparation of the experimental data. More specific, procedures are proposed to properly treat the preload in tensile and compression tests as well as to proof the consistency of experimental data from multiple experiments. Moreover, feasible formulations of the cost function for the parameter identification in terms of the stress measure, error type as well as order of the residual norm are studied and their effect on the fitting results is illustrated. After these preliminaries, invariant-based strain energy functions with decoupled dependencies on all three principal invariants are employed to identify promising models for each compound. Especially, appropriate parameter constraints are discussed and the role of the second invariant is analyzed. Thus, this contribution may serve as a guideline for the process of experimental characterization, data processing, model selection and parameter identification for existing as well as new materials. PubDate: 2023-01-09
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The intricacy of the real-world numerical optimization tribulations has full-fledged and diversely amplified necessitating proficient yet ingenious optimization algorithms. In the domain wherein the classical approaches fall short, the predicament resolving nature-inspired optimization algorithms (NIOA) tend to hit upon an excellent solution to unbendable optimization problems consuming sensible computation time. Nevertheless, in the last few years approaches anchored in nonlinear physics have been anticipated, announced, and flourished. The process based on non-linear physics modeled in the form of optimization algorithms and as a subset of NIOA, in countless cases, has successfully surpassed the existing optimization methods with their effectual exploration knack thus formulating utterly fresh search practices. Archimedes Optimization Algorithm (AOA) is one of the recent and most promising physics optimization algorithms that use meta-heuristics phenomenon to solve real-world problems by either maximizing or minimizing a variety of measurable variables such as performance, profit, and quality. In this paper, Archimedes Optimization Algorithm (AOA) has been discussed in great detail, and also its performance was examined for Multi-Level Thresholding (MLT) based image segmentation domain by considering t-entropy and Tsallis entropy as objective functions. The experimental results showed that among recent Physics Inspired Optimization Algorithms (PIOA), the Archimedes Optimization Algorithm (AOA) produces very promising outcomes with Tsallis entropy rather than with t-entropy in both color standard images and medical pathology images. PubDate: 2023-01-05
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The surge in internet use to express personal thoughts and beliefs makes it increasingly feasible for the social NLP research community to find and validate associations between social media posts and mental health status. Cross-sectional and longitudinal studies of social media data bring to fore the importance of real-time responsible AI models for mental health analysis. Aiming to classify the research directions for social computing and tracking advances in the development of machine learning (ML) and deep learning (DL) based models, we propose a comprehensive survey on quantifying mental health on social media. We compose a taxonomy for mental healthcare and highlight recent attempts in examining social well-being with personal writings on social media. We define all the possible research directions for mental healthcare and investigate a thread of handling online social media data for stress, depression and suicide detection for this work. The key features of this manuscript are (i) feature extraction and classification, (ii) recent advancements in AI models, (iii) publicly available dataset, (iv) new frontiers and future research directions. We compile this information to introduce young research and academic practitioners with the field of computational intelligence for mental health analysis on social media. In this manuscript, we carry out a quantitative synthesis and a qualitative review with the corpus of over 92 potential research articles. In this context, we release the collection of existing work on suicide detection in an easily accessible and updatable repository:https://github.com/drmuskangarg/mentalhealthcare. PubDate: 2023-01-03
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract For compressible flow simulations involving both turbulence and shockwaves, the competing requirements render it challenging to develop high-order numerical methods capable of capturing the discontinuities sharply and resolving the turbulence with high spectral resolution. In particular when deployed with the advanced large-eddy simulation (LES) approach, for which the governing equations are solved with coarse meshes, the solution is extraordinarily sensitive to the numerical dissipation resulting in large uncertainties for cross-code comparisons. Similar sensitivities have also been observed for a wide range of complex fluid predictions, e.g., turbulent reacting flows, two-phase flows, and transitional flows. In this paper, the family of high-order targeted essentially non-oscillatory (TENO) schemes on both the Cartesian and unstructured meshes is reviewed for general hyperbolic conservation laws with an emphasis on the high-speed turbulent flows. As a novel variant of popular weighted ENO (WENO) scheme, the TENO scheme retains the sharp shock-capturing capability of WENO and is suitable for resolving turbulence with controllable low numerical dissipation. The key success of TENO relies on a strong scale-separation procedure and the tailored novel ENO-like stencil selection strategy. In addition, the built-in candidate stencils with incremental width facilitate the construction of arbitrarily high-order (both odd- and even-order) schemes featuring superior robustness. Detailed performance comparisons between the WENO and TENO schemes are discussed comprehensively as well as the applications of TENO schemes to challenging compressible fluids. At last, the potential future developments to further boost the performance of TENO schemes from various perspectives are highlighted. PubDate: 2023-01-03
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Alzheimer's disease is an irreversible, progressive neurodegenerative disorder that destroys the brain and memory functionalities. In Alzheimer's disease, the brain starts shrinking, and over time it converts into dementia. The diagnosis of dementia takes an ample amount of time, around 2.8 to 4.4 years after the first clinical symptoms arise. Alzheimer's disease cannot be cured by any pharmacologic therapies (drugs) now on the market. Alzheimer's disease can only be avoided by early detection and prompt treatment. This paper proposes deep transfer learning models and MRI (Magnetic Resonance Imaging) images to detect the multiple stages of Alzheimer's disease such as "Very-Mild -Demented," "Mild-Demented," "Moderate-Demented," and "No-Demented." Data preprocessing and augmentation process are applied, enabling the model to detect the correct class of Alzheimer's disease. Then further deep transfer learning models (Resnet50, VGG19, Xception, DenseNet201, and EfficientNetB7) are used to classify and predict the early stages of Alzheimer's disease. It is observed that the DenseNet201 model performs the best, with a validation accuracy of 96.59%. The performance of Resnet50, VGG19, Xception, and EfficientNetB7 models was also recorded with validation accuracy of 93.52%, 95.08%, 89.77%, and 83.20%, respectively. The probability curve is then measured and the class-wise prediction of Alzheimer's disease is recorded using the area under curves and receiver operating curve (AUC-ROC) in order to analyze it more deeply. PubDate: 2023-01-03
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Distributed generation (DG) is a comprehensive, compact, and salient segment of the centralized power system due to its extensive prospective benefits in constrained/unconstrained power quality parameters. Traditional power plants have their own technical, environmental, and economic diminutions for the expansion of power generation in the distribution network. Furthermore, unreserved fossil fuel escorted the electricity companies towards the renewable resources of energy generation. The enhancement of power quality parameters, cost-saving, and fulfilling the energy demand can be achieved by the integration of optimized DG in the distribution network. This paper presents the conventional to modern mathematical approaches enacted for the objective variables’ empowerment via optimization of size and location of DG. The optimization intricacies, parameters, constraints, and benefits are also highlighted in this paper. The exhaustive data-based assessment carried out in this paper is a new work in the literature. PubDate: 2023-01-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Segmentation has been a rooted area of research having diverse dimensions. The roots of image segmentation and its associated techniques have supported computer vision, pattern recognition, image processing, and it holds variegated applications in crucial domains. To compile the vast literature on machine learning and deep learning-based segmentation techniques and proffer statistical, comprehensive, semi-automated, and application-specific analysis, which could contribute to the ongoing research. 16,674 studies have been filtered out from the pool of 22,088 studies collocated by executing a search string on the Scopus database. These studies are analyzed for their meta-data, comprehensive content and reviewed to identify key research areas using the topic modeling-based method (LDA). Also, the segmentation role for mathematical expression recognition has been fathomed out. IEEE is a ubiquitous name in the terms of the renowned publisher, reputed journal (IEEE Access), and most cited affiliation (#10,472). Three out of five extracted topic solutions by the LDA model be evidence of streaming research areas in image segmentation. Medical Image Processing, Machine Vision and Object Identification are the accentuated domains in the context. The streamlining of comprehensive analysis puts forth neural network-based approaches as a trend. Inquisition of segmentation techniques for mathematical expressions articulate neural-based segmentation techniques (CNN, RNN, LSTM) as preeminent segmentation techniques and geometrical features as focused features of the process. To sum up, the purpose of the current study is to summarize the best available research on image segmentation after synthesizing the results of an assorted set of studies. PubDate: 2023-01-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract We provide a publicly available collection of sixteen large-scale benchmark nonlinear state-space models in this contribution. The models are written in the MATLAB language and are scalable in spatiotemporal degrees of freedom. The aim is to provide the active research community with a suite of high-dimensional nonlinear models to test the state-of-the-art nonlinear model reduction strategies. We also review some of the most-widely employed reduction methods for these models. Furthermore, we also present some parametric nonlinear models to design and validate parametric model reduction schemes. PubDate: 2023-01-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract When disasters such as floods or earthquakes occur, we may not have a support of regular infrastructure based networks. This proves fatal because people who are trapped can not be easily located by search and rescue team. In such cases, airborne network consisting of miniaturized drones can be extremely beneficial in providing quick and effective coverage of the affected area, in an on-demand manner providing instant insights to rescue teams. While the challenges offered by such networks are plenty, the ongoing research and development shows promise to make such a technology more reliable and effective. In this paper, we discuss various disaster events in which network of drones can play a vital role in offering support to rescue operations. Mainly, the article discusses the protocols proposed by researchers for various layers of protocol stack including physical layer, data link layer, network layer, transport layer, application layer along with clustering protocols, time synchronization protocols and localization protocols. Finally, a brief summary of software simulation platforms and testbeds, along with future trends of Flying Ad-hoc networks have been provided. PubDate: 2023-01-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Acute leukemia is a cancer that starts in the bone marrow and is characterized by an abnormal growth of white blood cells. It is a disease that affects people all over the world. Hematologist study blood smears from patients to appropriately diagnose this anomaly. The methods used for diagnosis can be influenced by factors including the hematologist's experience and level of weariness, resulting in nonstandard results and even inaccuracies. The automatic detection of acute leukemia will produce robust results with precise accuracy. This systematic review gives a thorough investigation of the deep learning method for the classification and detection of acute leukemia. The systematic review adopted the PRISMA principle. Four online open source databases were utilized to find comparable articles, and a query featuring relevant keywords was created for the search purpose. Relevant publications were chosen from the search results based on inclusion and exclusion criteria to find answers to the four evolving research questions. The findings of the various studies were examined using the research questions that had been created.F1score and accuracy have been used as a performance matrix for the comparison purpose of CNMC and ALL IDB and self-acquired datasets. Consequently, various challenges faced by the authors have been highlighted. This systematic review article consists of a summary of the various automated detection and classification of acute leukemia in terms of four research questions. Different steps before classification like preprocessing, augmentation, segmentation, and feature extraction with various challenges faced by the author's different datasets and various challenges have been discussed in this paper. PubDate: 2023-01-01
Hybrid journal (It can contain Open Access articles)
[2467 journals]
