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.
Automatic Programming Assessment (APA) has been gaining lots of attention among researchers mainly to support automated grading and marking of students’ programming assignments or exercises systematically. APA is commonly identified as a method that can enhance accuracy, efficiency and consistency as well as providing instant feedback on students’ programming solutions. In achieving APA, test data generation process is very important so as to perform a dynamic testing on students’ assignment. In software testing field, many researches that focus on test data generation have demonstrated the successful of adoption of Meta-Heuristic Search Techniques (MHST) so as to enhance the procedure of deriving adequate test data for efficient t
... Show MoreThe study aimed to analyze the effect of meteorological factors (rainfall rate and temperature) on the change in land use in the marshes of the Al‐Majar Al‐Kabir region in southern Iraq. Satellite images from Landsat 7 for 2012 and Landsat 8 for 2022 were used to monitor changes in the land coverings, the images taken from the Enhanced Thematic Mapper Plus (ETM+) and Operational Land Imager (OLI) sensors of the Landsat satellite. Geometric correction was used to convert images into a format with precise geographic coordinates using ArcMap 10.5. The maximum likelihood classification method was used to examine satellite image data using a supervised approach, and the data were analyzed statistically. We obtained clear images of the area,
... Show MoreCarbonate reservoirs are an essential source of hydrocarbons worldwide, and their petrophysical properties play a crucial role in hydrocarbon production. Carbonate reservoirs' most critical petrophysical properties are porosity, permeability, and water saturation. A tight reservoir refers to a reservoir with low porosity and permeability, which means it is difficult for fluids to move from one side to another. This study's primary goal is to evaluate reservoir properties and lithological identification of the SADI Formation in the Halfaya oil field. It is considered one of Iraq's most significant oilfields, 35 km south of Amarah. The Sadi formation consists of four units: A, B1, B2, and B3. Sadi A was excluded as it was not filled with h
... Show MoreThe present study investigated the use of pretreated fish bone (PTFB) as a new surface, natural waste and low-cost adsorbent for the adsorption of Methyl green (MG, as model toxic basic dye) from aqueous solutions. The functional groups and surface morphology of the untreated fish bone (FB) and pretreated fish bone were characterized using Fourier transform infrared (FTIR), scanning electron microscopy (SEM) and Energy dispersive X-ray spectroscopy (EDS),respectively. The effect of operating parameters including contact time, pH, adsorbent dose, temperature, and inorganic salt was evaluated. Langmuir, Freundlich and Temkin adsorption isotherm models were studied and the results showed that the adsorption of basic dye followed Freundlich iso
... Show MoreThe present study investigated the use of pretreated fish bone (PTFB) as a new surface, natural waste and low-cost adsorbent for the adsorption of Methyl green (MG, as model toxic basic dye) from aqueous solutions. The functional groups and surface morphology of the untreated fish bone (FB) and pretreated fish bone were characterized using Fourier transform infrared (FTIR), scanning electron microscopy (SEM) and Energy dispersive X-ray spectroscopy (EDS), respectively. The effect of operating parameters including contact time, pH, adsorbent dose, temperature, and inorganic salt was evaluated. Langmuir, Freundlich and Temkin adsorption isotherm models were studied and the results showe
In this study, manganese dioxide (MnO₂) nanoparticles (NPs) were synthesized via the hydrothermal method and utilized for the adsorption of Janus green dye (JG) from aqueous solutions. The effects of MnO₂ NPs on kinetics and diffusion were also analyzed. The synthesized NPs were characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD), energy-dispersive X-ray analysis (EDX), and Fourier-transform infrared spectroscopy (FT-IR), with XRD confirming the nanoparticle size of 6.23 nm. The adsorption kinetics were investigated using three models: pseudo-first-order (PFO), pseudo-second-order (PSO), and the intraparticle diffusion model. The PSO model provided the best fit (R² = 0.999), indicating that the adsorpti
... Show MoreAbstract This study explores the extent to which public relations (PR) departments within Traqj governmental institutions are integrating artificial intelligence (AI) applications into their communication activities. The research adresses the growing importanc of AI in enhancing administrative efficieney, communication transparency, and stakeholder engagement. Adopting a descriptive research design, the study relied on an electtonic questionnaire distributed to PR profesionals across various ministries and government bodies, collecting 100 valid responses. The indings reveal that while younger PR practitioners are actively embracing AI, older employees show limited engagement. Most participants acquired AI-related skills through self- learn
... Show MoreChemical pollution is a very important issue that people suffer from and it often affects the nature of health of society and the future of the health of future generations. Consequently, it must be considered in order to discover suitable models and find descriptions to predict the performance of it in the forthcoming years. Chemical pollution data in Iraq take a great scope and manifold sources and kinds, which brands it as Big Data that need to be studied using novel statistical methods. The research object on using Proposed Nonparametric Procedure NP Method to develop an (OCMT) test procedure to estimate parameters of linear regression model with large size of data (Big Data) which comprises many indicators associated with chemi
... Show MoreSurvival analysis is one of the types of data analysis that describes the time period until the occurrence of an event of interest such as death or other events of importance in determining what will happen to the phenomenon studied. There may be more than one endpoint for the event, in which case it is called Competing risks. The purpose of this research is to apply the dynamic approach in the analysis of discrete survival time in order to estimate the effect of covariates over time, as well as modeling the nonlinear relationship between the covariates and the discrete hazard function through the use of the multinomial logistic model and the multivariate Cox model. For the purpose of conducting the estimation process for both the discrete
... Show More