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.
Distribution of light intensity in the flat photobioreactor for microalgae cultivation as a step design for production of bio-renewable energy was addressed in the current study. Five sizes of bioreactors with specific distances from the main light source were adopted as independent variables in experiential design model. The results showed that the bioreactor’s location according to the light source, determines the nature of light intensity distribution in the reactor body. However, the cross-section area plays an important role in determining the suitable location of reactor to achieve required light homogeneity. This area could change even the expected response of the light passing through the reactor if Beer-Lambert's law is adopted.
... Show MoreMobile Wireless sensor networks have acquired a great interest recently due to their capability to provide good solutions and low-priced in multiple fields. Internet of Things (IoT) connects different technologies such as sensing, communication, networking, and cloud computing. It can be used in monitoring, health care and smart cities. The most suitable infrastructure for IoT application is wireless sensor networks. One of the main defiance of WSNs is the power limitation of the sensor node. Clustering model is an actual way to eliminate the inspired power during the transmission of the sensed data to a central point called a Base Station (BS). In this paper, efficient clustering protocols are offered to prolong network lifetime. A kern
... Show MoreThe problem of finding the cyclic decomposition (c.d.) for the groups ), where prime upper than 9 is determined in this work. Also, we compute the Artin characters (A.ch.) and Artin indicator (A.ind.) for the same groups, we obtain that after computing the conjugacy classes, cyclic subgroups, the ordinary character table (o.ch.ta.) and the rational valued character table for each group.
Ag nanoparticles were prepared using Nd:YAG laser from Ag matel in distilled water using different energies laser (100 and 600) mJ using 200 pulses, and study the effect of the preparation conditions on the structural characteristics of and then study the effect of nanoparticles on the rate of killing the two types of bacteria particles (Staph and E.coli). The goal is to prepare the nanoparticle effectively used to kill bacteria.
Nowadays, most of the on-chip plasmonic single-photon sources emit an unpolarized stream of single photons that demand a subsequent polarizer stage in a practical quantum cryptography system. In this paper, we numerically demonstrated the coupling of the light emitted from a quantum emitter (QE) at 700 nm wavelength to the propagation mode supported by an on-chip hybrid plasmonic waveguide (HPW) polarization rotator. Our results proved that the light emitted is linearly polarized at 0º, 45º/−45º, and 90º with propagation lengths of 5 μm, 3.3 μm, and 3.9 μm, respectively. Moreover, high power-conversion efficiency was obtained from an applied transverse magnetic (TM) mode (0º-polarization) to a transverse electric (TE) (90º-polari
... Show MoreThe present work presents design and implementation of an automated two-axis solar tracking system using local materials with minimum cost, light weight and reliable structure. The tracking system consists of two parts, mechanical units (fixed and moving parts) and control units (four LDR sensors and Arduino UNO microcontroller to control two DC servomotors). The tracking system was fitted and assembled together with a parabolic trough solar concentrator (PTSC) system to move it according to information come from the sensors so as to keep the PTSC always perpendicular to sun rays. The experimental tests have been done on the PTSC system to investigate its thermal performance in two cases, with tracking system (case 1) and without trackin
... Show MoreThe present study deals with the websites of Iraqi political parties on the internet to identify the effectiveness in providing communicative applications that help audience to participate, express their opinions, their positions, and other aspects reflecting the extent of employing modern technological tools to allow opportunities for political, and democratic participation since the internet has become an effective tool for political communications of political parties. The research sample includes eight political parties. The research concludes that the Iraqi political parties do not employ interactive communication patterns to reflect their interests in communicating with the public, providing opportunities for their participation an
... Show MoreThe introduction of Industry 4.0, to improve Internet of Things (IoT) standards, has sparked the creation of 5G, or highly sophisticated wireless networks. There are several barriers standing in the way of 5G green communication systems satisfying the expectations for faster networks, more user capacity, lower resource consumption, and cost‐effectiveness. 5G standards implementation would speed up data transmission and increase the reliability of connected devices for Industry 4.0 applications. The demand for intelligent healthcare systems has increased globally as a result of the introduction of the novel COVID‐19. Designing 5G communication systems presents research problems such as optimizing