Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.
Densities ρ and viscosities η for several concentrations of amino acids (Serine, Cysteine and Threonine) at different temperatures (298.15, 303.15 and 308.15K) have been measured. On the basis of these data, the apparent molal volumes v , partial molal volumes at infinite dilution v , slope Sv , Gibbs free energy of activation for viscous flow of solution ∆G1,2 and Jones – Dole Bcoefficients were calculated the nature of solute-solvent and solute-solute interactions have been discussed in terms of the values of v , v , Sv and B-coefficents
Praise be to God, who created people from one soul, and made her husband, and prayers and peace be upon His Messenger, Mercy given and the grace given, and on a machine and companions pure. Divorce cases have increased in recent years in a worrying manner, especially since divorce has unforeseen consequences at the individual and social levels. The source of concern stems from the fact that the cohesion and integrity of society starts from a family as it is the cornerstone of the social structure, which is the foundation upon which the large society is based. Marital life may suffer from problems and obstacles that are difficult to solve, so the man presents the divorce of his wife and that the phenomenon of divorce is not a result of it, b
... Show MoreAbstract
This research aims to improve the provided health service level inside Baghdad hospitals and the Yarmouk educational, as well as to shed light on the reality of the health service and the quality within the major operations room in both hospitals, as the operations room represent the research community, as was the use of some quality tools Pareto and Ishikawa diagram to measure and assess the level of quality provided, and include research problem to find out what are the problems and obstacles facing the process of improving quality in both hospitals, and whether there are scientifically accurate method to assess the quality of health service in Baghdad's Yarmouk hospital and educational . Where the researcher h
... Show MoreSoils that cause effective damages to engineer structures (such as pavement and foundation) are called problematic or difficult soils (include collapsible soil, expansive soil, etc.). These damages occur due to poor or unfavorited engineering properties, such as low shear strength, high compressibility, high volume changes, etc. In the case of expansive soil, the problem of the shrink-swell phenomenon, when the soil reacts with water, is more pronounced. To overcome such problems, soils can be treated or stabilized with many stabilization ways (mechanical, chemical, etc.). Such ways can amend the unfavorited soil properties. In this review, the pozzolanic materials have been selected to be presented and discussed as chem
... Show MoreRecurrent strokes can be devastating, often resulting in severe disability or death. However, nearly 90% of the causes of recurrent stroke are modifiable, which means recurrent strokes can be averted by controlling risk factors, which are mainly behavioral and metabolic in nature. Thus, it shows that from the previous works that recurrent stroke prediction model could help in minimizing the possibility of getting recurrent stroke. Previous works have shown promising results in predicting first-time stroke cases with machine learning approaches. However, there are limited works on recurrent stroke prediction using machine learning methods. Hence, this work is proposed to perform an empirical analysis and to investigate machine learning al
... Show MoreThe main objective of resources management is to supply and support the site operation with necessary resources in a way to achieve the required timing in handing over the work as well as to achieve the cost-realism within the budget estimated. The research aims to know the advantage of using GIS in management of resources as one of the new tools that keep pace with the evolution in various countries around the world also collect the vast amount of spatial data resources in one environment easily to handled and accessed quickly and this help to make the right decision regarding management of resources in various construction projects. The process of using GIS in the management and identification of resources is of extreme importance in t
... Show MoreRoaming data is an important source of information about the political and social activities of a country. And this is true for Iraq situation after 2003 when the mobile companies started their business. In this paper, data of subscribers roamed onto foreign networks (inbound roamers) is collected; it consists of the name of the Radio Control Point, the counter dealing with this type of information, Mobile Network Code/Mobile Country Code tupel. This data is processed. Results obtained out of this process show the classification of the inbound roamers (according to their countries) during the occupation period (2008-2009). These results reflect the political situation of Iraq at that time. Information resulted from this proc
... Show More