Search

In an era where artificial intelligence (AI) is rapidly shaping the future of technology, it is becoming increasingly integrated and utilized across various fields, unlocking a wide range of applications and use cases. Among these, the field of cybersecurity will benefit significantly, as AI has the potential to enhance traditional methods of detection and prevention. This paper aims to present the latest advances in AI technology, exploring how AI can be synergistically connected with cybersecurity and highlighting the types of applications already developed in this area. Additionally, this paper will discuss the ethical considerations and current challenges associated with the use of AI to ensure a more secure environment.

As artificial intelligence (AI) is becoming increasingly integrated into critical infrastructures, it brings about both transformative benefits and unprecedented risks. AI has the potential to revolutionize the efficiency, reliability, and responsiveness of essential services, but it can also offer these benefits along with the vulnerability to a growing array of sophisticated adversarial attacks. This paper explores the evolving landscape of adversarial threats to AI systems, highlighting the potential of nation-state actors to exploit these vulnerabilities for geopolitical gains. A range of adversarial techniques is examined, including dataset poisoning, model stealing, and privacy inference attacks, and their potential impact on sectors such as energy, transportation, healthcare, and water management is assessed. The consequences of successful attacks are substantial, encompassing economic disruption, public safety risks, national security implications, and the erosion of public trust. Given the escalating sophistication of these threats, this paper proposes a comprehensive security framework that includes robust incident response protocols, specialized training, the development of a collaborative ecosystem, and the continuous evaluation of AI systems. The findings of this study 11 underscore the critical need for a proactive approach to AI security in order to safeguard the future of critical infrastructures in an increasingly AI-driven world.

The research aimed to explore the potential of advanced machine learning (ML) algorithms in clinical and biomedical research. The significance of frameworks like generative adversarial networks (GANs), autoencoders, and autoregressive models in tackling the issues of representation learning and the quality of generated content is highlighted. This paper also presents a proposed system architecture for integrating generative artificial intelligence (GenAI) into healthcare processes. This architecture encompasses components for data ingestion, preprocessing, model training, image enhancement, diagnostic analysis, and user interfaces for healthcare providers and patients, utilizing advanced artificial intelligence (AI) models. The paper underscores the necessity of robust data governance frameworks, ethical guidelines, and secure infrastructures to mitigate the associated risks. By fostering collaborative AI-human systems and continuously assessing ethical implications, the healthcare industry can fully exploit GenAI's potential to improve patient outcomes and operational efficiency.

Energy consumption prediction in buildings is a major issue in the field of energy efficiency and resource management. In recent decades, the use of Artificial Intelligence (AI) has become an increasingly common approach to improve the accuracy and reliability of these predictions. In this paper, the most used digital technologies such as: AI, Big Data, Internet of Things (IoT), Blockchain, Cloud computing and 5G, which can fundamentally transform the way energy consumption predictions are made in buildings, are presented; the different types of data used in energy consumption prediction are analysed, such as weather data, energy use data and building characteristics; the various AI methods and algorithms are presented, e.g. neural networks, decision trees, support vector and machine learning algorithms, used to improve prediction accuracy. This paper focuses on a conceptual energy prediction model developed for the project "Intelligent system for predicting energy consumption in buildings (PRECONERG)", a project in the first stage of development.

This paper investigates the utilisation of ChatGPT (Generative Pre-trained Transformer) in the field of Facial Emotion Recognition (FER), streamlining development processes and making them smoother and faster. With ChatGPT, coding and debugging happen fast, leading to the swift creation of performing models in mere minutes. It highlights the transformative potential of integrating ChatGPT into image processing workflows, accelerating progress in computer vision, graphics, and multimedia. This paper also shows the importance of word choice and developer skill in finding the right balance between speed and accuracy. It sets the stage for further investigations into advanced Large Language Models (LLMs) and finetuning parameters, leading towards faster, more efficient image processing techniques.

Data is being stored electronically increasingly more, and the importance of databases has therefore increased significantly. Given that the data collected is used for various reasons and the sensitivity of the data can differ, the importance of database security is crucial. Although, database security has been on the “agenda” of security experts for a long time, technologies have changed and databases have evolved accordingly. In this paper, the aim is to offer a refreshed perspective of the security measures implemented in databases nowadays, with a comparison study of two popular databases: Oracle and Microsoft SQL. The comparison will provide the strengths and weaknesses of both database systems after analyzing the authentication, authorization, encryption, auditing and user management features provided by both, to help determine which Database Management System has a more comprehensive asset of tools to protect its internal data. Furthermore, this paper can also offer a guide for database administrators, to help them choose the best strategy and tools to secure their databases.

The Metaverse concept describes a virtual collective space where users engage with simulated or computer-generated environments, enabled by cutting-edge technologies such as Virtual Reality (VR), Augmented Reality (AR), Blockchain, Artificial Intelligence (AI), and 5G Networks. This convergence allows the creation of complex virtual worlds, enhances user interactions, enables dynamic environments, facilitates seamless integration across platforms, and ensures scalability and persistence, alongside several privacy, security, ethical and legal challenges to be considered. The constant advancement of these technologies is pivotal in shaping a future where the line between physical and virtual realities blurs, unlocking limitless possibilities in realms such as communication, entertainment, education, and business. This leads to novel opportunities for exploration and user engagement. This paper aims to present an overview of the leading technologies driving the transition from the current Internet to the Metaverse, enabling users to interact naturally and intuitively with digital objects and environments. The study explores well-established interactive simulation technologies like virtual reality and augmented reality, along with emerging technologies like somato-sensory interfaces, holographic imaging, and brain-computer interaction interfaces. Emphasizing their potential, the paper demonstrates how these technologies can offer users unparalleled levels of involvement in interactive activities spanning gaming, socializing, education, and business.

Natural Language Processing (NLP) is a branch of computer science, artificial intelligence, information engineering, and linguistics that focuses on enabling machines to understand, interpret, and generate human language. Machine translation, sentiment analysis, speech recognition, and question-answering systems are just a few examples of NLP applications. NLP has advanced rapidly in recent years, owing to the availability of large amounts of text data and the development of deep learning models, which has resulted in increased accuracy and efficiency in NLP tasks. This paper tackles the state of the art of NLP, its methodology and techniques, and its applications in different fields. The paper also discusses the role of NLP to improve education.

This paper presents a scientific approach regarding the existing technologies and methodologies for establishing the reputation of an Internet domain, as well as some considerations related to the introduction of artificial intelligence techniques for the dynamic monitoring of the domains. The work is part of a research project whose purpose is to ensure the degree of trust given to domains against malicious activities in the Internet space by automatically monitoring them and offering solutions to keep data safe. The huge number of newly registered domains, the proliferation of malicious domains that represent attack vectors for DNS servers are the main causes that require finding an automated solution for establishing the reputation level of each domain. By establishing the level of reputation of the domain/domains owned, their future users (authorities, state institutions, private companies, individuals, etc.) will have a correct image of the degree of trust of the used domain, thus, creating a safer Internet space. Along with the classical techniques for establishing infested domains (blacklist) currently used by DNS administrators, the automatic detection tools presented in the paper have the advantage of fast detection and automatic updating of possible compromised domains.

This paper aims at formulating standardization recommendations, both for Romanian specialists developing telehealth software and platforms and for Romania as EU member, which is in a national landscape with a multitude of unconnected medical systems and own architectures that no longer have their place in a common European medical reality. The paper addresses a range of issues such as communication between devices, international recognition of national systems, elimination of proprietary API’s with mandatory open interfaces, choice of a common software development language, the usage of a common European cloud, semantic interoperability, ensuring security and respecting patient privacy. All of these cannot be understood and implemented without the knowledge of the European framework, the European ethics guidelines, the strategy for electronic health records, so that the interactions between healthcare systems can be made totally on all levels (data, services and processes) at national level. This system, in turn, needs to be integrated into the single European digital health system.

This paper seeks to improve the efficiency of a whole wind power conversion system (WPCS) based on a dual star asynchronous machine (DSAM) incorporated with a flywheel energy storage system (FESS) based on an induction machine (IM). The FESS is connected to the wind generator through the DC bus link in order to control the amount of power delivered to the analysed network. The FESS has gained a lot of attention recently because of its low costs, long lifetime, and clean technology. This paper proposes a powerful control strategy based on a synergetic approach with a vector control applied to the DSAM and a Fuzzy logic controller applied to the FESS machine, this control can ensure a quick response time and a good stability of the system over the whole operating range, and can guarantee system robustness. Moreover, the Particle Swarm Optimization algorithm is proposed for fixing the problem of synergetic controller parameter tuning, which enhances the effectiveness of the control system. The dynamic model of the suggested system of the wind power generator is simulated in Matlab/Simulink for all points of operation. The obtained results which are presented and discussed in this paper, show how well the proposed model worked.

Some features of left and right brained people can be determined using the left and right brain dominance theories. It can assist in the development of training syllabus for brain-balancing education topics. When performing any action, human focus or concentration is essential. This paper will examine the concentration levels of patients with left and right brain dominance using electroencephalogram (EEG) data. Brain activity may be tracked and recorded using EEG waves. The human brain’s thinking and attention cause brain waves to alter in distinct frequency bands. A frequency-based EEG signal can be cleaned up and characteristics can be extracted using the baseline correction method. As a result, the EEG Topographical Power Spectral Density Value is created. The main objective of this paper is to compare the concentration levels of people with different brain dominance. Inversely, EEG signal can be used to predict whether a person has a left or a right brain dominance.

Information Technology (IT) is critical to strengthening corporate governance. Thus, organizations require a well-designed and implemented IT governance framework. This paper aims to evaluate the maturity level of IT governance in Saudi Arabia, via the COBIT IT governance framework. Three processes were investigated: defining a strategic IT plan; assessing and managing IT risks; and project management. A questionnaire was developed, based on the COBIT maturity model, and distributed among IT workers at different levels. Results demonstrate that the private sector is the most mature within the investigated areas; more development is required for the government sector and even the semi-government sector. The paper explores different points of strength and weakness in terms of IT governance maturity in different sectors. Moreover, it presents steps that may improve the IT governance maturity level in government and semi-government sectors. This could lead to more future research and investigation in different variables that may affect the development and maturity of IT governance.

For a large number of citizens dealing with public authorities can feel like a harrowing experience. With a large amount of bureaucracy and ever-changing rules and procedures, it can be quite confusing for people to achieve their goals. The aim of this paper is to prove that a chatbot can be trained with minimal coding experience, by having a robust dataset created by people with domain-specific knowledge. This would empower every public institution to customise a helper chatbot. This chatbot would not only provide common information about required documents for whatever needs to be done, but also more general information pertaining to the specific aspects of that institution, such as its location, how to get to it, or public transportation connections nearby. The aim is to provide the necessary steps so that any public institution willing to undertake this digital transformation step can do so without much outside help or technical know-how in the field of Artificial Intelligence.

The last decade has witnessed the exponential growth of Virtual Reality (VR) as an affordable, powerful, and disruptive technology in a variety of fields, among which, notably, are healthcare and well-being. Particularly for the elderly people, VR has already been demonstrated, by a plethora of experiments and studies, as extremely effective in supporting a multitude of therapeutical interventions, from neuromotor and cognitive recovery and maintenance to mental health and wellbeing and socialisation. However, due to the novelty of the technology, the scientific and industry communities are missing both an overarching perspective covering all these intervention areas, as well as a common understanding of the results, challenges, limitations, possibilities, specificity, best practices, design approaches, and integration with traditional therapies, for each area. This hinders the development of commercial-level solutions and widespread adoption in clinical guidelines. This paper aims to cover these gaps, by identifying and analysing five of the most relevant VR intervention areas for elderlies, providing an overview and examples, for each area, and concluding with cross-area perspectives and pinpoints of the most needed research topics, to support the widespread adoption of VR-based interventions for elderly healthcare and wellbeing.

This paper aims to analyze the classic approach towards the Workplace ecosystem and make a set of general recommendations, business-agnostic which applied, allow organizations to reach a Digital and Integrated Workplace. An overview is presented on the key components of a reliable IT department, with a focus on the critical infrastructures across all levels: Front-End, Middleware, and Back-End systems. It starts by gathering the most important innovation methods in the IT domain. Then, our research follows on a concrete example of a company that underwent the implementation of these changes and check how effective they have proven to be. The research focuses on the adoption of these measures and how they has been reflected in their employee satisfaction level. Finally, the most interesting Key Performance Indicators from multiple standpoints are advocated: resource allocation (cost, time, manpower), user experience, and the impact of emerging technologies based on AI and service governance. While it is clear that all of these riders' implementation depends mainly on the budget, according to Gartner, 62% of IT services buyers confirmed that their organizations are willing to pay an extra 11% to 20% of the already allocated budget for top premium IT services, tailored to their custom needs with a focus on the user experience.

The public administration constitutes the mechanism through which government policies and public programs are implemented for citizens. Coherent communication and the quality of services provided are extremely important and must take place in a safe and collaborative environment, both for citizens and the organizations with leadership role involved. The rapid development of emerging technologies has determined a huge innovation potential in the virtual environment, with results especially in the economic, educational and health fields. This paper aims to highlight the potential of Blockchain technology and the Metaverse space in the field of public administration in the coming years, starting from the motivation that of all categories of citizens must benefit from high quality administrative services in a fully immersive world. The technological and application aspects of Blockchain technology addressed in this research can support the digitization of public services and the development of clear governance models. Potential public services in the Metaverse have been suggested along with the role of Blockchain technologies as an alternative to traditional government services.

The current paper establishes a rapid, real-time algorithm for the synthesis of the reaction matrices used by compensators that reduces multivariable linear systems to extended systems that are cyclical. Thus, the driving of the extended systems (original multivariable system plus compensator) is reduced to a reaction that depends on the outputs and a set of references calculated by the proposed algorithm. The algorithm derives the structure of the compensator and the values of its components (amplifiers and integrators) as well as the values of the references. The paper is divided into two main sections: the derivation of the algorithm and the description of the program.

The aim of this paper is to analyze and evaluate three different forms of Maximum Power Point Tracking (MPPT) techniques under standard and various atmospheric conditions. In the literature, different MPPT methods are proposed and studied by researchers, let us quote the well known Perturb & Observe (P&O), the incremental conductance, fuzzy logic based MPPT, neural networks based MPPT,etc. The major drawback found is the effectiveness of these techniques under sudden variations in weather conditions such as irradiance and temperature. Another drawback of some MPPT techniques is the response time which is important to achieve the optimum power value. In this sense, optimization of P&O method becomes the objective of our work, to track the point of maximum power in a photovoltaic system under different weather conditions. In this paper, we will improve the P&O technique which makes it possible to react quickly to sudden changes in solar radiation and to follow the maximum power point with very low oscillations. The main idea is to use a PID regulator to make the voltage of the PV panel much closer than the MPP voltage, this method is named indirect P&O. The simulation based comparative study showed that the MPPT indirect P&O technique led to better performance.

Automated recognition and classification of aquatic species (fish, shrimp etc.) are very useful for studies dealing with the count of species for population evaluation, fish behaviour analysis, monitoring of the ecosystem and understanding the association between species and the ecosystem. Transformers have shown phenomenal success in computer vision problems. However, it demands extensive data for classification tasks. Existing traditional vision transformers necessitate large datasets for heightened accuracy, perpetuating the belief that transformers are data-hungry. This paper aims to dispel this idea by introducing the Amended Dual Attention oN Self-locale and External (ADANSE) mechanism-based vision transformer for classifying underwater (fish) species. In this approach, input images undergo block-tokenization, followed by the application of the proposed attention mechanism, Amended Dual Self Locale and External attention. The Amended dual self-locale attention layer extracts deep feature representations and the external attention mechanism considers the potential relationship among all image blocks. Then, the outputs from both attention mechanisms are further feeding the Multi-Layer Perceptron (MLP) network for species recognition. A proprietary fish database on complex environments is acquired and a self-collected fish database is constructed. This includes the species of Penaeus vannamei, Hypostomus plecostomus, Oreochromis niloticus and its juvenile. When compared to existing ViT networks, the proposed ADANSE network proved to perform better, attaining an accuracy of 90.9% on proprietary datasets and 92% on standard benchmark datasets, emphasising its robust performance even on small-sized images. This highlights the potential of the ADANSE ViT network to address data dependency concerns and achieve competitive accuracy levels in underwater species classification.

This paper presents a comparative analysis of web-based machine learning models, specifically examining Google Vertex AI, Google Teachable Machine, Azure Machine Learning and Salesforce Einstein Vision. The objective is to assess their suitability for integration into a medical information system as a classification module for medical images. The comparative evaluation considers factors such as model accuracy, ease of integration and scalability. The findings aim to guide the selection of an optimal machine learning solution for enhancing the capabilities of medical image classification within a healthcare context.

As the production of electronics increases exponentially each year, the need for them to be able to function properly (according to the documentation) is critical. The functional verification process is a complex and time consuming step when delivering a proper System-On-a–Chip, which makes it a great candidate for different AI implementations and try-outs. In functional verification the UVM Methodology, where UVM stands for Universal Verification Methodology is vastly used combined with a hardware verification language such as System Verilog. Writing various, elaborated testbenches for difficult architectures such as RDMA (Remote Direct Memory Access), USB (Universal Serial Bus), etc. and achieving 100% functional coverage is laborious. Here, Generative AI may be able to help to improve this process by a considerable amount. This paper aims to present the different trials, aspects and challenges of producing an APB UVM testbench with Generative AI, using ChatGPT.

The proliferation of Internet of Things (IoT) devices introduces significant challenges in network security and scalability. Software-Defined Networking (SDN) emerges as a promising framework to offer more flexible and intelligent network management capabilities. This paper delves into the integration of SDN principles within an IoT environment to bolster security and scalability. Through a proof-of-concept deployment utilizing the Floodlight controller and Mininet-WiFi, the approach involves the collection and analysis of data on average latency, packet loss, throughput, and jitter to evaluate network performance. Additionally, Python scripts and Wireshark are employed for an in-depth network analysis. The findings illustrate that SDN integration can adeptly manage the augmented network load, evidenced by minimal increases in latency and packet loss, while maintaining acceptable throughput and jitter levels. Furthermore, the Floodlight controller's capability to identify and counter Distributed Denial of Service (DDoS) attacks underscores its potential to enhance IoT network security. The results affirm that SDN can significantly elevate the scalability and security of IoT networks, presenting a viable solution to manage the escalating demands of IoT deployments. Future endeavors will aim to extend the network scale and investigate alternative SDN controllers to substantiate the scalability of the conclusions.

This paper proposes a robust model predictive control method for a class of linear discrete time, uncertain and disturbed systems. A relationship between the system disturbance, the states and control input exists, which is used to remove, through several manipulations, the disturbance from the control optimization problem. Moreover, in comparison with other several previous studies, the disturbance does not act directly on the system, a system disturbance matrix being introduced. In principle, the main objective is to find a control law by solving a min-max problem in which a robust performance objective is to be minimized. Instead, an equivalent optimization problem is solved and an upper bound is found for the robust performance objective using a Lyapunov function. With the upper bound, the equivalent control optimization problem is formulated. The solutions of the equivalent optimization problem are used to construct the control law. A Matlab simulation, using Yalmip toolbox, indicates that the states are stabilized to zero and the control input tends to zero.

Service providers in 4G wireless communication networks are constantly challenged to accommodate more customers within a limited allotted capacity. OFDM (Orthogonal Frequency Division Multiplexing) is used to boost the data rate of a wireless medium while reducing interference. It is important to analyse the performance of the FFT- and DWT-OFDM by using different methods and algorithms. In this paper FFT and DWT is employed with different algorithms such as Cooley-Tukey, Prime factor, Haar and Daubechies. These algorithms have been modeled and analysed for the designed software model. A Comparative analysis has been carried out by considering SNR (Signal-to-Noise Ratio) and BER (Bit Error Rate) values for QAM modulation. MATLAB simulation showed that DWT-based OFDM system performed much better than the FFT-based OFDM systems on AWGN channel. 5G technology involves communication engineering and embedded system engineering to enhance the required data rate for different applications. The proposed system implementation with the 5G MIMO-OFDM on Zynq UltraScale provides a data rate within a range of 10 Gbps to 50 Gbps for the different order of the combinations of FFT/IFFT or DWT/IDWT and QAM. The different orders of the combinations were achieved through programmability and reconfigurability. The performance of the DWT-OFDM is more complex and computationally intensive than that of FFT-OFDM in hardware. In future wok the proposed system could be improved so as to become a multi-user MIMO(MU-MIMO).

In cryptography, a mode of operation is the way of processing plaintext and encrypted text blocks within a block cipher algorithm, or it is the presentation of a method of chaining blocks in a block cipher. Several models exist with their strengths, some are more vulnerable than others, and some combine authentication and security concepts. In this paper, a new mode of operation is proposed, inspired by the braid of long hair which we call mode CBLCH (Cipher Braided Long Hair Chaining). The HILL cipher will be used to validate it and to compare it with ECB (Electronic Code-Book: code dictionary) and CBC mode (Cipher Block Chaining: sequence of blocks) by the influence of a pixel change on the plaint-image and the encrypted image.

Internet of Things (IoT) generates huge amount of data, that needs to communicate between the IoT enabled devices. These communications are vulnerable to security attacks and are malicious enough to cause harm to connected devices. The invasive communication and security breaches have to be identified and should be dealt with in order not to cause further damage and consequences. The objective of this work is to distinguish intentional communications from insecure communications between the IoT devices. The intentional communications can be different from the insecure communications in their patterns. Artificial intelligencebased machine learning approaches have the technologies to identify patterns of the intentional or insecure communications. In this paper, Random Forest, Decision Tree, SVM and 1DCNN have been used to discriminate patterns belonging to intended and unintended messages. To evaluate this technique, IoT-23 dataset is used, the proposed machine learning based approach obtaining a performance of 99.25% accuracy with the benchmark dataset. The proposed approach is compared with the state-of-the-art methods. It is observed that the proposed Random Forest method outperforms the existing ones with sufficient patterns to identify. To enhance the performance of the poorly performing classifiers on the imbalanced dataset, a potential solution to be applied on this dataset is also explored and proposed.

The current study describes a novel approach for controlling a Permanent Magnet Synchronous Motor (PMSM) that utilizes intelligent speed sensorless control based on Field-Oriented Control (FOC), by taking inter-turn short-circuit (ITSC) faults into account and by using a Luenberger observer and adaptive fuzzy logic. The speed loop control in conventional indirect vector control systems typically uses a traditional PI regulator with fixed gains, which may not always be enough to achieve the desired control performance. To address this limitation, a Fuzzy Logic Controller (FLC) is proposed for speed control in the field-oriented control FOC of PMSM. This approach is designed to handle the challenges posed by continuous variations in machine parameters, model uncertainties, nonlinear dynamics, and external disturbances. Additionally, the need for a rotor shaft position sensor can lead to disadvantages such as increased cost, weight and noise and reduced reliability. To overcome these issues, the speed and rotor position are estimated using a Luenberger observation, without requiring a separate sensor. In order to evaluate the efficacy of intelligent controller proposed in the present paper, digital simulations were conducted using the MATLAB&SIMULINK, under varying operating conditions. These included sudden changes in reference speed, load torque, and varying degrees of failure for healthy and degraded states. The simulation results demonstrated that the adaptive FLC outperformed the conventional PI controller in robustness, against parameter variation and stator faults. Additionally, the controller was less sensitive to external load torque disturbances and exhibited a fast dynamic response.

The purpose of this paper was to investigate the relationship among technology readiness, the implementation of artificial intelligence-based customer relationship management, and the promotion of the corporate social responsibility of B2B companies active in Iran's auto parts supply industry. This study has an applicative nature and is based on the descriptive survey research method. The statistical population sample for this study included 247 people, all of them managers and senior experts at the analysed companies with at least 7 years of work experience. A questionnaire was applied for collecting research data, and its validity and reliability were confirmed through the combined reliability and Cronbach's alpha indices. After collecting the necessary data in order to test four research hypotheses, the structural equation modelling method was employed by using the SmartPLS software. The results for testing the first research hypothesis showed that technology readiness has a positive effect on corporate social responsibility and that it directly predicts 0.810 (81%) of the changes related to corporate social responsibility in the context of the analysed companies. Also, the analysis of the results obtained for testing the fourth research hypothesis showed that technology readiness through artificial intelligence-based customer relationship management has a positive effect on corporate social responsibility and that technology readiness indirectly predicts 0.493 (49.3%) of the changes related to corporate social responsibility. Based on this, it can be acknowledged that the analysed companies can take steps towards improving corporate social responsibility by increasing technology readiness, strengthening their communication technology infrastructure and also by improving customer relationship management capabilities through artificial intelligence.

Aquaculture is the practice of reproducing, increasing and yielding aquatic organisms, such as aquatic animals and aquatic plants, in confined water bodies, like ponds, lakes, rivers, oceans, etc. Fishery represents one of the activities of aquaculture. The key function affecting fisheries management is the fishing activity. Operations like locating and counting fish are used to enhance this practice. There is a strong demand for underwater fish identification for multiple uses in sustainable fisheries. Real time monitoring helps to improve fishing activities. Deep Learning Techniques are used to train the computer with the available existing image data, faster GPUs, and algorithms employed to detect, locate and classify various objects within an underwater image or video with high accuracy. A well-liked object detection model, namely YOLO (You Only Look Once), is renowned for its quickness and precision. This paper presents a state-of-the-art version of the YOLO model for detecting and counting fish from underwater images or videos. The primary objective is to develop a system for automatic fish detection using an advanced convolutional neural network YOLOv8 and compare the results with the ones of the YOLOv7 model. It is proved that YOLOv8s performs better than YOLOv7, as YOLOv8s achieves a mAP@0.5 of 0.964, a precision of 0.929 and an IoU of 0.7.

Artificial Intelligence (AI) has led to advancements in multiple fields of research, and music has always been a field of high interest. Music is an important part of life and various studies have shown the link between better living and listening to music. From completing melodies to composing music from scratch, there are many applications of AI in this domain. This paper aims to analyse one such application of using AI for music generation, specifically for instrumental accompaniment. Instrumental accompaniment is essentially the instrumental music that is composed to support or complement a melody. Creating instrumental accompaniment for music generally requires extensive musical knowledge or forming a band together with skilled instrumentalists. Build Your Own Band (BYOB) attempts to simplify this process with the help of AI. In this research work, three transformer models are employed for training various accompanying instruments. Here, the proposed transformer model accepts a melody line as input and produces an accompanying track with instruments like bass instruments, the guitar and string instruments. One of the main challenges was to make sure that these instruments produce a cohesive sound.

This paper gives a clear visual appearance of the speed predictive control by a Model Reference Adaptive System (MRAS) observer for a Permanent-Magnet Synchronous Motor which is the heart of the electric vehicle’s powertrain, The proposed approach used is Field Oriented Control (FOC) of the PMSM drive in case of speed sensor failure and the importance of substituted it by MRAS observer system to ensure the continuity of precaution situation drive and even get rid of its problems at all, thus improve the reliability of electric vehicle powertrain system and securing drivers and pedestrians from dangers. The performances and robustness tests of the traction chain are achieved by real scenarios (driving cycles) such as NEDC, SFTP US06, and JP-15 combined with the vehicle dynamic model. This effectiveness study of (PMSM) drives enhanced by an observer in an electric car application confirms the high estimation accuracy and robustness of torque control of the proposed sensorless approach in several real road situations.

On social media, the way identity is presented is a misrepresentation of reality, which does not set the proper framework and opens the way for Incel ideology and other equally severe societal problems. Essentially, there is a fine line between having a place to vent emotionally and developing your identity and encouraging beliefs and attitudes that society does not want. The first one is positive, while the second one risks causing harm to others, both physically and emotionally. The anonymity of the internet can promote an aggressive and sometimes borderline violent environment when opinions clash. In fact, the online environment is now a more effective and globalised way of spreading extremist messages and targeting disaffected and marginalised young people to enlist them in violent causes. In fact, the online environment is now a more effective and globalised way of spreading extremist messages and targeting disaffected and marginalised young people to enlist them in violent causes. This article aims to provide an informed understanding of the Incel ideology's scope and nature. The paper considers how the Incel ideology has evolved, including its changing perspective, to demonstrate how members of this community can be a factor in the rise of cybercrime.

While online shopping has become popular in recent years, certain problems related to customer expectations about the appearance and fitting of clothing items still need to be solved. Online customers looking for clothing products may be unsatisfied as they do not have the possibility to try on clothes and view how they would look on them. Additionally, vendors are burdened with the cost of items returned by unsatisfied customers. Augmented Reality (AR) can help vendors to attract more customers and enhance their shopping experience. Customers are more likely to be satisfied with their purchases when they can try the clothes on before purchasing them, which ultimately reduces the return costs for vendors. This paper presents an AR mobile application called Smart Fitting which was developed in order to allow online customers to try on clothing items by using 3D models virtually. The most recent AR development tools for iOS operating systems, including XCode and Unity, are utilized for creating the proposed mobile application. After an integration testing for the mobile application functions, seven target female consumers participated in the usability evaluation for this mobile application which was meant to prove its ease of use and evaluate the customers’ satisfaction. The obtained results showed that Smart Fitting can improve the users shopping experience by providing a more realistic clothes visualization than the traditional online shopping interfaces.

The last few years have brought about a tremendous increase in the speed of technological advancement and digitalisation which led to an exponential uptick in the amount of data that is being either stored or moved across the cyberspace. This data, either under the ownership of a company operating in the private sector, or of a public sector organization, has become a priority target for malicious actors. From actors that operate under the alleged guise of financially motivated reasons such as ransomware operators to nation state threat groups and politically motivated actors that engage in hacktivist campaigns, an important certainty is that confidential data has become a top priority target for the various types of malicious actors operating across the cyber threat spectrum. Whether it is being sold to the highest bidder, released for free or used in the context of nation state affairs, confidential data became a high-value resource which, in combination with the increasingly complex and politically charged international context of the last few years, has brought about an uptick in number of cyber operations that could be classified as espionage. The aim of this paper is to present a study showcasing the various ways in which data theft, as the main objective, can materialize through malicious cyber campaigns ranging from ransomware operations executed by documented and well-known actors, to malware operations executed under the radar by obscure malicious actors.

Data analytics and AI have become extremely relevant in today’s football landscape. Data is benefiting clubs in gaining a competitive advantage on and off the field by empowering them to harvest information for improving player performance, decreasing injuries, and increasing commercial efficiency. Tracking data in football, that is the data of x, y coordinates of all 22 players on the pitch every second adds a lot of value in terms of a team’s decision-making and recruitment strategy. Via the use of this data, each player’s decision-making ability is also measured. The players and the ball in each frame are identified using YOLOv5, which returns their coordinates. These detections are then passed to DeepSORT, which assigns IDs to each player and keeps track of the frame by frame, by feeding each player detection to the model. The K-Means model is employed to determine the jersey color of the players to identify the two teams. Finally, the detected coordinates are multiplied with the homogeneous matrix computed using the Sports Camera Calibration through the Synthetic Data paper approach to accomplish the perspective transform. The mathematical model hypothesized and implemented uses pitch control and expected threat to assess each player’s decision-making ability, which will be a strive to enhance the recruiting of players.

Sensorless control has become an attractive and important topic from an industrial perspective. The presence of sensors in electronic speed drives can reduce their robustness and reliability while increasing costs and mounting complexity. Therefore, mechanical sensors will be replaced by a calculation algorithm that provides speed estimation from the electrical terminal of the machine. It is imperative that the performance required from this estimator be close to or similar to that provided by a physical sensor. In this paper, different estimation and observation techniques will be presented. Then, Sensorless vector control based on a mutual – MRAS estimator and Fuzzy mutual – MRAS estimator for Dual Star induction Motor (DSIM) to estimate the speed, as well as the resistances for both the stator and rotor will be introduced. Finally, the results of the simulation will be shown and discussed.

This paper presents a taxonomy and evaluation of models for forecasting and estimation of various solar irradiation components received at ground; the aim is to compensate for the lack of measures in meteorological stations throughout the country. The chosen models have been proposed in the literature by scientists and researchers. These models are a function of the main meteorological parameters such as relative humidity, temperature, pressure, as well as geographical parameters of the site considered as longitude, latitude and altitude, they also depend to some astronomic parameters like the day number and the solar constant. Three models were studied and simulated for the aim to estimate the solar radiation components at ground level. Finally, comparisons are carried out between the obtained results through simulation and measured values actually found in some Algerian sites where such measurements are possible and available. The aim is to establish the criteria for selecting models according to the geographic and meteorological conditions. The obtained results through simulation of the three models, give good estimate of solar components where errors between the calculated values and those measured are negligible. We can say that the chosen models are good for estimation of solar radiation at ground.

The main purpose of the present paper is to build a robust synergestic control of the permanent magnet synchronous machine (PMSM), that drives a flywheel to form an energy storage system that is integrated into a hybrid PV-wind system. The biggest inconvenience of the exploitation of clean sustainable sources like solar and wind energies is the continuous variation of the weather conditions, appearing as a frequent change in the wind currents (wind speed) and solar radiations. This problem causes a fluctuation in electricity production by the wind turbines and photovoltaic panels, which is a serious drawback particularly in the isolated standalone system when the system is incapable of generating demanding power. Thus, the use of hybrid system with more than one source, especially when it is supported by an energy storage system like FESS, is an effective solution for this drawback. It is able to re-generate power when the hybrid PV-wind system cannot generate enough electricity. Moreover, the synergetic control guarantees fast response, asymptotic stability of the closed-loop system in the all-range operating condition, and system robustness in presence of parameter variation. Computational simulation results confirm that the proposed method consistently outperforms other techniques and proves its effectiveness and benefits of using such a system.

Home health monitoring of the elderly has grown in recent years and the context of pandemic has facilitated its evolution, offering the opportunity for medical service providers to expand and improve the quality of eHealth systems and products. A major market potential is represented by the field of eHealth – in this case the home health monitoring of the elderly, the context of the pandemic being a decisive factor in the exploration of this potential. The need to design viable solutions geared towards a target market becomes an inevitable challenge to improve profitability and therefore stakeholder value. The paper presents in detail a SWOT analysis highlighting the key points of eHealth monitoring systems, identifying the optimal potential of these solutions. The case study presents a viable eHealth solution highlighting a competitive advantage in personalised home monitoring for the elderly as well as a financial projection.

Surfacing weld represents the deposition of a filler material on the surfaces or edges of a base metal. The objective of the process is to improve the mechanical characteristics of metallic structures that are subjected to cyclic loads. Heat transfer occurs at the interface between the welded beads and the base metal. Thus, the structural performances of the final product are limited. In this regard, the evaluation of the temperature distribution by means of simulation tools can provide essential information to the designers of such structures. The present paper proposes a new approach for the estimation of the thermal gradients that occur during the welding process. In the first stage, a black box system is developed with the support of experimental data. A simulation approach that is based on the Finite Element Method is proposed for the transient analysis of the heat transfer. The conductance and capacitance of the model are defined based on the parameters of the “Black box” system.

Sentiment analysis is concerned with determining whether a certain material contains online information which expresses positive or negative sentiments. The tools for performing this analysis should be able to identify and assess thoughts and feelings with a reasonable degree of accuracy on feelings that are made openly available by people. It is expected that sentiment analysis would be performed for social media. That is why this paper investigates online social media, as sentiment analysis has become an important subject, and it is one of the approaches employed in the field of natural language processing. Sentiment analysis was applied for an Arabic Twitter dataset in order to identify the feelings expressed by the textual tweets and determine whether they were positive, negative, or neutral. Bigrams and unigrams were used when employing the multinomial Naïve Bayes, Gaussian Naïve Bayes, Logistic Regression, and Support Vector Machines (SVM) machine learning algorithms. The Logistic Regression algorithm achieved the highest accuracy, that is with 63.40%. The Long Short-Term Memory (LSTM) neural network was used for the deep learning-based classification, and it reached an accuracy rate of 70%, a figure which proved to be higher than the results shown in the related works.

Healthcare informatics emphasizes on the gathering, processing, assessment, and utilization of health records, information, and expertise. By connecting genetics, proteomics, and patient records, bioinformatics uses computational methods and tools to research and analyze huge biological datasets, completely understand illness, and grasp genetics and proteomics. Interpreting genomic and proteomic data for basic biological research, as well as for use in healthcare, drug discovery, and related fields, is the main focus. Large amounts of electronic data collection in the healthcare industry lead to the development of analytics. Data analytics is adept at enhancing healthcare, cutting costs, and preserving people. The explosion of data leads to the use of data analytics in the healthcare industry to explore understandings and make smart judgments. In order to create a better health service, this paper focuses on bioinformatics, healthcare informatics, and analytics. It examines the advantages, how each one helps to improve the health service, how bioinformatics, health informatics, and analytics connect, and lastly, what the coming times holds for healthcare informatics as well as analytics.

In recent years, the open science and, specifically, the open access to scientific publications and research data have been topics of interest for the international scientific community, including the European Union and its member states. Recently, this interest has been amplified by the development and launch of the Horizon Europe framework program and national research and innovation programs. For our country, the National Strategy for Research, Innovation and Intelligent Specialization 2022-2027, recently adopted, includes the two topics among its specific objectives. In this context, the paper analyzes the results obtained by the Horizon 2020 program in the implementation of the open access policy for publications and research data generated by its own projects, as highlighted in the final program monitoring report. The provisions of the model grant agreement regarding open access and the pilot dedicated to open research data are presented. The typology of monitoring indicators is detailed and the main indicators and their values are listed. Some conclusions drawn from the accumulated experience, with the aim of improving this activity in Horizon Europe, are reviewed. For this program, the focus on impact-based evaluation of its results is also highlighted. This approach is illustrated by short-, medium- and long-term impact monitoring indicators specific to open science.

Nowadays, the vehicle routing problem is one of the most important combinational optimization problems and it has received much attention because of its real application in industrial and service-related contexts. It is considered an important topic in the logistics industry and in the field of operations research. This paper focuses on the comparison between two metaheuristics namely the Genetic Algorithm (GA) and the Discrete Artificial Bee Colony (DABC) algorithm in order to solve the vehicle routing problem with a capacity constraint. In the first step, an initial population with good solutions is created, and in the second step, the routing problem is solved by employing the genetic algorithm which incorporates genetic operators and the discrete artificial bee colony algorithm which incorporates neighbourhood operators which are used for improving the obtained solutions. Experimental tests were performed on a set of 14 instances from the literature in the case of which the related number of customers ranges typically from 50 to 200, in order to assess the effectiveness of the two employed approaches. The computational results showed that the DABC algorithm obtained good solutions and a lower computational time in comparison with the GA algorithm. They also indicated that the DABC outperformed the state-of-the-art algorithms in the context of vehicle routing for certain instances.

Coronavirus disease 2019 (COVID-19) has caused nearly 600 million individual infections worldwide and more than 6 million deaths were reported. With recent advancements in deep learning techniques, there have been significant efforts to detect and diagnose COVID-19 from computerized tomography (CT) scan medical images using deep learning. A retrospective study to detect COVID-19 using deep learning algorithms is conducted in this paper. It aims to improve training results of pre-trained models using transfer learning and data augmentation The performance of different models was measured and the difference in performance with and without using data augmentation was computed. Also, a Convolutional Neural Network (CNN) model was proposed and data augmentation was used to achieve high accuracy ratios. Finally, designed a website that uses the trained models where doctors can upload CT scan images and get COVID-19 classification (https://covid-e46e8.web.app/) was designed. The highest results from pre-trained models without using data augmentation were for DenseNet121, which was equal to 81.4%, and the highest accuracy after using the data augmentation was for MobileNet, which was equal to 83.4%. The rate of accuracy improvement percentage after using data augmentation was about 3%. The conclusion was that data augmentation could improve the accuracy of COVID-19 detection models as it increases the number of samples used to train these models.

In this paper, a fuzzy sliding mode controller (FSMC) with adaptive gain is proposed for a class of MIMO nonlinear systems. The coupled system can be divided into two different subsystems, each of which can be separated into two sliding surfaces created by the state variables. The integration of these two sliding surfaces requires the introduction of an intermediary variable. In order to ensure that the control system is stable, the adaptive gain is derived using the Lyapunov method. The synthesized fuzzy sliding mode control with adaptive gain is applied to an inverted pendulum to illustrate the suggested control technique. The performance and robustness of the proposed controller are compared, terms of time, with those of provided by fuzzy sliding mode controller. This comparison reveals the superiority of the proposed method over the fuzzy sliding mode controller, in terms of the system stabilization and tracking of the trajectory.

Currently, the Brushless Dual-Fed Induction Machines (BDFIMs) are very often used in Wind Energy Conversion Systems (WECSs). The features of BDFIM are its high reliability due to its brushless operation, and the related lower capital and operational costs. This paper presents a field-oriented control algorithm for a Brushless Doubly-Fed Induction Generator (BDFIG). The proposed decoupled control is determined on the Power-Winding (PW) stator-flux frame and can be employed for controlling the rotation speed and its reactive power. In order, to avoid the use of an anemometer with speed sensors and to maximize wind energy extraction. Maximum Power Point Tracking (MPPT) based on Optimal Torque Control (OTC) is proposed. Several simulation results under different operating conditions are provided in order to prove the effectiveness of the proposed algorithm.

Mobile Ad-hoc Networks (MANETs) are a kind of networks comprised of nodes which are mobile from place to place in nature, hence the name is MANET. Being an ad-hoc network, it does not need a centralised control infrastructure. Along with the mobile nodes, these type of networks also deals with the radio propagation models in which the nodes communicate with one another using different propagation models. These propagation models have their own kind of communication scenarios in which direct Line of Sight (LoS), Non-Line of Sight (N-LoS), two ray ground reflection etc. By communicating with one another, these propagation models undergo issues like going away from each other in which the distance increases and it directly affects their efficiency. In this paper, two-ray ground and free-space radio waves propagation models have been taken into consideration to check the distance-based scenarios in different distance ranges. With the help of Ad-hoc On-demand Distance Vector (AODV) routing scheme the simulations have been performed in NS-2 by taking the performance evaluation metrics that are average throughput, average latency and average packet drop. The simulations revealed that distance has a direct impact on the performance of propagation models. It has been concluded that two ray ground gives the best performance in contrast with the free space propagation model. The main reason is that two ray ground uses the two-way ground reflection in which it communicates via LOS and N-LOS. The simulations results have revealed that the proposed radio wave propagation models have a huge impact with the distance-based scenario to evaluate the performance of AODV routing scheme.

The purpose of this article was to highlight the sentiment analysis for specific Arabic tweets related to the COVID-19 Worldwide Epidemic. The technique proposed in this paper focused on using the machine learning algorithm with the purpose of applying sentiment analysis on a dataset which contained 4,575 Arabic tweets on the COVID-19 pandemic while also employing the Logistic Regression and Naive Bayes algorithms as classifiers for comparing the achieved results between them. This study showed the suitability and efficiency of a system using machine learning models for the analysis of Arabic tweets. The experimental outcomes revealed that the highest accuracy was reached by employing the Logistic Regression algorithm”, namely, 97%”. Twitter is one of the most widely used gateways of social media for the people who want to express their opinions and emotions. This study contributes to highlighting the task of sentiment analysis for the Arabic tweets about the COVID-19 pandemic by predicting the people’s awareness about the Coronavirus in the Arab World.