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.
The term "nano gold," also known as "gold nanoparticles," is commonly used. These particles are extremely small, with a diameter of less than 100 nm, which is only a fraction of the width of a human hair. Due to their tiny size, nano gold particles are often found in a colloidal solution, where they are suspended in a liquid stabilizer. This colloidal gold is essentially another name for nano gold. The main method for producing gold nanoparticles in a colloidal solution is the citrate synthesis technique, which involves combining different solutions to precipitate the gold nanoparticles. In biological systems, copper complexes play a significant role at the active sites of many metalloproteins. These complexes have potential applications in
... Show MoreThe objective of the research is to identify the efficiency of risk management in various names at Baghdad International Airport in the face of various risks (financial - technical - human - natural ..) facing the sample of the search of the General Establishment of Civil Aviation and the Iraqi Airways Company where the researcher identified the hypothesis that summarizes There is a significant significant correlation between risk management, risk management and risk review and assessment. The researcher used the means of research from observation and interviews with the relevant officials in this field, as well as used the questionnaire and distributed a sample of 170 employees in the field of risk management (SMS Department) in Iraqi A
... Show MoreThis paper presents a nonlinear finite element modeling and analysis of steel fiber reinforced concrete (SFRC) deep beams with and without openings in web subjected to two- point loading. In this study, the beams were modeled using ANSYS nonlinear finite element
software. The percentage of steel fiber was varied from 0 to 1.0%.The influence of fiber content in the concrete deep beams has been studied by measuring the deflection of the deep beams at mid- span and marking the cracking patterns, compute the failure loads for each deep beam, and also study the shearing and first principal stresses for the deep beams with and without openings and with different steel fiber ratios. The above study indicates that the location of openings an
In this paper, a new class of nonconvex sets and functions called strongly -convex sets and strongly -convex functions are introduced. This class is considered as a natural extension of strongly -convex sets and functions introduced in the literature. Some basic and differentiability properties related to strongly -convex functions are discussed. As an application to optimization problems, some optimality properties of constrained optimization problems are proved. In these optimization problems, either the objective function or the inequality constraints functions are strongly -convex.
The importance of the research from a practical point of view lies in the fact that it presents a set of statistics and data that give a clear picture of how the Iraqi newspapers (the subject of the study) deal with the visit of "Pope Francis" to Iraq ، and what are the most prominent indicators and manifestations of that visit in promoting societal peace among the Iraqi public. From a scientific point of view، the research provides another scientific addition to the media library، especially with regard to journalistic treatments and methods of framing the Arab international press for the subject of the visit، which could be a starting point for other researchers to complete qualitative research in this field. The research prob
... Show MoreMH Hamzah, AF Abbas, International Journal of Early Childhood Special Education, 2022
Abstract:
Research Topic: Ruling on the sale of big data
Its objectives: a statement of what it is, importance, source and governance.
The methodology of the curriculum is inductive, comparative and critical
One of the most important results: it is not permissible to attack it and it is a valuable money, and it is permissible to sell big data as long as it does not contain data to users who are not satisfied with selling it
Recommendation: Follow-up of studies dealing with the provisions of the issue
Subject Terms
Judgment, Sale, Data, Mega, Sayings, Jurists
Particulate matter (PM) emitted from diesel engine exhaust have been measured in terms of mass, using
99.98 % pure ethanol blended directly, without additives, with conventional diesel fuel (gas – oil),to
get 10 % , 15 %, 20 % ethanol emulsions . The resulting PM collected has been compared with those
from straight diesel. The engine used is a stationary single cylinder, variable compression ratio Ricardo
E6/US. This engine is fully instrumented and could run as a compression or spark ignition.
Observations showed that particulate matter (PM) emissions decrease with increasing oxygenate
content in the fuel, with some increase of fuel consumption, which is due to the lower heating value of
ethanol. The reduction in
The study aimed to prepare a measure of metacognitive thinking commensurate with learning the skill, preparing educational units using the Claus Meyer model for metacognitive thinking, and learning the skill of defending the court in volleyball. To identify the effect of educational units using the model (and Claus Meyer) for metacognitive thinking and learning the skill of defending the court in volleyball. The two researchers used the experimental approach with the design of the experimental and control groups.The research community consisted of students of the second stage / College of Physical Education and Sports Sciences / University of Baghdad for the academic year 2021-2022, whose number is (385) st
... Show More