ABSTRACT: 

A program has been developed that utilises machine learning methodologies for the classification of gas discharge images of liquid solutions. The glow patterns of liquid solution droplets in an electromagnetic field were recorded using the gas discharge visualization (GDV) method. The method utilises a mass-produced Bio-Well device, does not require consumables, and acquires images in approximately one minute. In the developed algorithm, classifiers were trained to demonstrate high accuracy in distinguishing various types of water, including tap-filtered water, distilled water, water from three different springs, tea with sugar, and solutions with magnesium, salt, and shungite impurities. The developed approach is employed in the research of water purification methods using electromagnetic fields.

Keywords: Gas Discharge Visualization camera (GDV); Machine Learning; Computer Vision; Image Processing.

ABSTRACT: 

The rapid advancement of Information Technology (IT) has significantly altered the business landscape, leading to both opportunities and challenges for traditional businesses. As digital tools and platforms become increasingly integral to modern commerce, traditional business models face growing pressure to adapt or risk obsolescence. The aim of this study was to investigate the effects of IT integration on local businesses, focusing on its transformative potential and associated challenges. With a sample size of 80 businesses, the research aimed to explore how IT adoption influenced operational efficiency, revenue growth, customer interaction, and workforce productivity. The study utilized a mixed-methods approach, combining quantitative data on IT integration levels and impacts with qualitative insights from interviews. Findings revealed that while IT adoption led to significant improvements in business operations and revenue for some businesses, others faced notable challenges, including high costs and technical issues. The research highlighted that businesses with comprehensive IT integration often experienced enhanced customer engagement and increased productivity, whereas those with minimal or no IT integration struggled to achieve similar benefits. The study concluded that strategic planning, adequate training, and addressing technical and financial barriers are crucial for businesses to maximize the advantages of IT and effectively navigate the disruptions it presents.

Keywords:  Information Technology (IT), Traditional Businesses, Digital Transformation, Operational Efficiency, Revenue Impact, Customer Engagement, IT Integration, Technological Disruption, Business Adaptation.

ABSTRACT: 

The availability of internet services and the abundance of data sources, such as news websites, scientific publications, blogs, forums, and social media networks, have resulted in vast data repositories. Navigating through these repositories in search of interesting information presents a significant challenge. One effective solution to this challenge is automatic text summarization, which identifies the most important content by condensing the original text into a new form known as a summary. In this study, we develop a deep learning-based Arabic text summarization approach utilizing a Restricted Boltzmann Machine (RBM). The RBM is trained using summaries created based on human experience to identify the optimal weights that serve as discriminative factors for the importance of text features. The EASC dataset is employed to evaluate the proposed approach, alongside the ROUGE (Recall-Oriented Understudy for Gisting Evaluation) toolkit. For a fair comparison, several related methods from the literature are also included. The proposed approach achieves higher scores in Recall, Precision, and F-measure for the ROUGE-2 metrics compared to the referenced methods. The superior performance of the proposed deep learning approach validates the effectiveness of using human-generated summaries as a source of human experience for training the RBM.

Keywords:  automatic text summarization, Restricted Boltzmann Machine, deep learning, human experience.

ABSTRACT: 

Computational fluid flow modeling is as important to the geotechnical engineer as field study because it provides a improved considerate of how the water and soil interact.  Due to many influences such as weather variation, oil extraction, and building construction, the technical engineer must understand the nature of the soil, its properties, and its water content. This article reviews the most important classical and modern approaches to modeling fluid flow through soil, starting with soil types and properties and ending with modern modeling trends, especially now in light of the development of artificial intelligence. Three computational methods have been prominently reviewed in the imitation of the fluid flows investigations in geotechnical engineering: Finite Element Method (FEM), Finite Difference Method (FDM), and computational fluid dynamics (CFD) not only because of their popularity but also because of the accuracy of the results extracted by their use. Also in this study, a review comparison between stability analysis and numerical methods in soil-water modeling. This study concludes that the ways in which water and soil interact must be continually reviewed, particularly in light of developments in other sciences such as engineering, physics, artificial intelligence, and advanced methods for managing big data.

Keywords: Geotechnical Engineering, Fluids Dynamics, Fluid Flow Modelling, Soil–Water Interaction.