Iraqi Journal For Computer Science and Mathematics

Homotopy analysis method for solving fuzzy Nonlinear Volterra integral equation of the second kind

Rana Hussein — 2024-08-06

This paper introduces the homotopy analytic HAM, an effective and reliable method for solving computationally challenging fuzzy Volterra nonlinear integral equations of the second type. Numerical examples are provide to demonstrate the accuracy of HAM. To check for existence and uniqueness, the study employs the Banach fixed point theory and homotopy analysis. Finally the study resolves our issue using the MAPLE programme.

Embedded Deep Learning to Improve the Performance of Approaches for Extinct Heritage Images Denoising

Alaa HamzaOmran — 2024-08-11

Many advanced deep convolutional neural network (DCNN) methods have proven their efficacy in reconstructing the texture of super-resolution images (SR) from low-resolution images (LR). Nevertheless, the objective of achieving super-resolution (SR) reconstruction using Deep Convolutional Neural Networks (DCNN) becomes difficult when the input image is distorted by noise. Photographs captured at the inception of the camera are presently regarded as a cultural heritage that chronicles an important period in human history; however, they are marred by low resolution and noise as a result of obsolescence and the primitive nature of the technology that captured them, in contrast to the technological advances that cameras benefit from today. We proposed embedded deep learning to improve the performance of approaches for extinct heritage images denoising to denoise and reconstruct Baghdad heritage images. First, stage super-resolution (SR) noisy image generation from low-resolution (LR) heritage noisy image aims to enable to extraction of noise features for the target images. Second, remove visible noise features on images and restore their surface texture, giving them a more modern and clearer scene while preserving their original identity. PSNR, SNR, and SSIM quantitative metrics and the visual comparison analysis between the proposed method and state-of-art methods: Total variation denoising (TV), Bilateral filter denoising (BF), Median filter Denoising (MF) Gaussian filter image denoise (GF), and Non-Local Bayes (NLB) denoising demonstrated a better performance in reducing noise from the target images and a high-frequency flow with more information. Our approach restores heritage in a way that mimics modern photographic scenes using deep learning algorithms.

Resource Allocation and Edge Computing for Dual Hop Communication in Satellite Assisted UAVs enabled VANETs

Mustafa Maad Hamdi — 2024-06-13

VANETs are highly attractive and is used in maximum of the applications of cross-regional communication. To increase the coverage of the vehicular network, Unmanned Arial Vehicles (UAVs) are introduced, and they get connected with the satellite networks to perform heterogeneous communication. With the help of this connectivity, the communication quality of ground level to air medium is increased. Currently the vehicle usage is highly increased and as a results of communication link failure, improper resource allocation are arises whither abruptly assumes a stability about a network with that increases an energy consumption and communication delay in the heterogeneous networks. In these conditions, thus study is idea of Resource Allocation and Edge Computing for Dual Hop Communication (RAEDH) in introduced in satellite assisted UAVs enabled VANETs. The major sections of the approach are UAV assisted mobile computing, resource allocation among the vehicles and the UAVs, and dual communication among the vehicles and the UAVs.Through these methods the input resources are properly allocated and that reduces the power utility and communication delay. Initially, the vehicular network is established, incorporating trusted components like TA, RSU, and CRS. Subsequently, mobile edge computing reduces energy consumption through computation offloading and optimized UAV trajectory selection. Resource allocation, facilitated by whale optimization, ensures effective utilization across vehicles. The implementation of this method is done in NS3, and the scenario is analyzed using two parameters like number of vehicles and its speed. The output parameters that remain thought-out over a performance examination stay throughput, end-to-end delay, energy efficiency, packet loss, packet delivery ratio, and routing overhead, and as well those results are compared with the earlier methods. Finally, dual-hop transmission between vehicles and UAVs enhances delivery ratio and throughput. From the results and discussion, it has been proven that the proposed RAEDH-SAVs attained maximum results in terms of energy efficiency, delivery ratio, and throughput.

Unlocking the Potential: Synergizing IoT, Cloud Computing, and Big Data for a Bright Future

zeena alkateeb — 2024-06-10

The convergence and integration of the Internet of Things (IoT), Cloud Computing (CC), and Big Data (BD) offer huge potential for transformative progress that will support the massive industrial revolution that is so prevalent in today’s digital landscape. This prompts an exploration of the synergies among these three domains and investigates how their integration can unlock new opportunities for a brighter future. This integration seamlessly connects billions of IoT devices, leverages the potential of CC, and efficiently manages vast datasets generated by these devices. The amalgamation of BD analytics with IoT and CC empowers organizations to extract invaluable insights, foster data-driven decision-making, and fuel innovation across diverse industries. Nevertheless, the incorporation of these systems also gives rise to notable difficulties, encompassing issues such as data protection, worries about privacy, the ability to scale, and the management of data. The paper delves into these challenges, explores strategies, and examines the hurdles and best practices to address them effectively through a comprehensive examination of the potential benefits, challenges, and mitigation strategies. In addition, this paper offers insights into how synergizing IoT, CC, and BD can pave the way for a brighter, more promising future in technology and society.

A Comprehensive Review on Cybersecurity Issues and Their Mitigation Measures in FinTech

Guma Ali — 2024-06-10

The fourth industrial revolution has seen the evolution and wide adoption of game-changing and disruptive innovation, "financial technologies (FinTech), around the globe. However, the security of FinTech systems and networks remains critical. This research paper comprehensively reviews cybersecurity issues and their mitigation measures in FinTech. Four independent researchers reviewed relevant literature from IEEE Xplore, ScienceDirect, Taylor & Francis, Emerald Insight, Springer, SAGE, WILEY, Hindawi, MDPI, ACM, and Google Scholar. The key findings of the analysis identified privacy issues, data breaches, malware attacks, hacking, insider threats, identity theft, social engineering attacks, distributed denial-of-service attacks, cryptojacking, supply chain attacks, advanced persistent threats, zero-day attacks, salami attacks, man-in-the-middle attacks, SQL injection, and brute-force attacks as some of the significant cybersecurity issues experienced by the FinTech industry. The review paper also suggested authentication and access control mechanisms, cryptography, regulatory compliance, intrusion detection and prevention systems, regular data backup, basic security training, big data analytics, use of artificial intelligence and machine learning, FinTech regulatory sandboxes, cloud computing technologies, blockchain technologies, and fraud detection and prevention systems as mitigation measures for cybersecurity issues. However, tackling cybersecurity issues will be paramount if FinTech is to realize its full potential. Ultimately, this research will help develop robust security mechanisms for FinTech systems and networks to achieve sustainable financial inclusion.

New Structures of Soft Permutation in Commutative Q-algebras

Moataz Sajid Sharqi — 2024-07-15

A Generating Distorted CAPTCHA Images Using a Machine Learning Algorithm

Saba Abdulbaqi Salman — 2024-07-31

CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart)
have become universal in web security systems to differentiate between automated bots and human users. This
research presents a novel approach for generating and classifying distorted CAPTCHA images utilizing machine
learning techniques. The process involves developing a random text and rendering it onto an image, introducing
distortion for security. The proposed method involves developing CAPTCHA images by combining text rendering
and controlled distortion techniques. These images are then utilized to train a random forest classifier for accurate
recognition. A Random Forest classifier is employed to recognize the generated CAPTCHA images. Experimental
results demonstrate the approach's efficacy in achieving high validation accuracy. The validation accuracy of the
classifier demonstrates its effectiveness in deciphering distorted images. Thus addressing the challenge of creating
CAPTCHAs that are both human-readable and resistant to automated recognition.

Improving security in the 5G-based medical Internet of Things to improve the quality of patient services

israa albarazanchi — 2024-07-23

The Internet of Medical Things (IoMT) is like a tech upgrade that benefits patients by reducing healthcare
costs, making medical care more accessible, and improving the quality of treatment. To make IoMT devices smart and capable,
they need super-fast 5G support. However, there are security concerns when using IoMT devices that can put a patient's data
and privacy at risk. For instance, someone could eavesdrop on your medical data due to weak network access management and
data encryption. Many systems use encryption methods to protect data, but these methods often fall short when it comes to the
high security standards required for healthcare data and patient service quality. In our research, we introduce a new solution
called the Hybrid MD5 and Threefish Encryption (HMTE) to make IoMT more secure and improve the quality of care for
patients. To ensure efficient use of energy, we employ a smart approach when choosing a cluster head. When it comes to
sending data, we use the Trust-Based Energy Efficient Routing Protocol (TEERP). We carefully evaluate different aspects like
cost, encryption and decryption speed, and the level of security while analyzing our proposed method. We also compare our
solution with existing methods. Our data shows that our recommended solution outperforms existing methods, particularly in
terms of enhancing security to improve the quality of care for patients.

Development of synthetic data generator for ornament based on data mining techniques

Saad Ahmed Dheyab — 2024-08-31

Artificial intelligence tools depend on generating various images on the available datasets, but there is a lot of data that is not completely available, especially images of heritage and archaeological ornament. In this paper, a tool was developed to generate ornament based on the geometric shapes of heritage ornaments. A group of basic geometric shapes was collected. The new ornament are reshaped in three stages, relying on the permutation equation and reshaping of the overlapping sizes of the shapes, In the first stage, the number of layers that make up the ornament is determined. In the second stage, the formation of the ornament is determined through rotation or mirroring. In the final stage, the set of ornaments is generated in the form of an integrated matrix.

Convex Optimization Techniques for High-Dimensional Data Clustering Analysis: A Review

Ahmed Yacoub Yousif — 2024-07-31

Clustering techniques have been instrumental in discerning patterns and relationships within
datasets in data analytics and unsupervised machine learning. Traditional clustering algorithms struggle to handle
real-world data analysis problems where the number of clusters is not readily identifiable. Moreover, they face
challenges in determining the optimal number of clusters for high-dimensional datasets. Consequently, there is a
demand for enhanced, adaptable and efficient techniques. Convex clustering, rooted in a rich mathematical
framework, has steadily emerged as a pivotal alternative to traditional techniques. It amalgamates the strengths of
conventional approaches while ensuring robustness and guaranteeing globally optimal solutions. This review offers
an in-depth exploration of convex clustering, detailing its formulation, challenges and practical applications. It
examines synthetic datasets, which serve as foundational platforms for academic exploration, emphasizing their
interactions with the semi-smooth Newton augmented Lagrangian (SSNAL) algorithm. Convex clustering provides
a robust theoretical foundation, but challenges, including computational limitations with expansive datasets and
noise management in high-dimensional contexts, persist. Hence, the paper discusses current challenges and
prospective future directions in the domain. This research aims to illuminate the potency and potential of convex
clustering in modern data analytics, highlighting its robustness, flexibility and adaptability across diverse datasets
and applications.

Healthcare Privacy-Preserving Federated Transfer Learning using CKKS-Based Homomorphic Encryption and PYHFEL Tool

Anmar A. Al-Janabi — 2024-08-11

Digitization of healthcare data has shown an urgent necessity to deal with privacy concerns within
the field of deep learning for healthcare organizations. A promising approach is federated transfer learning,
enabling medical institutions to train deep learning models collaboratively through sharing model parameters rather
than raw data. The objective of this research is to improve the current privacy-preserving federated transfer
learning systems that use medical data by implementing homomorphic encryption utilizing PYthon for
Homomorphic Encryption Libraries (PYFHEL). The study leverages a federated transfer learning model to classify
cardiac arrhythmia. The procedure begins by converting raw Electrocardiogram (ECG) scans into 2-D ECG
images. Then, these images are split and fed into the local models for extracting features and complex patterns
through a finetuned ResNet50V2 pre-trained model. Optimization techniques, including real-time augmentation
and balancing, are also applied to maximize model performance. Deep learning models can be vulnerable to
privacy attacks that aim to access sensitive data. By encrypting only model parameters, the Cheon-Kim-Kim-Song
(CKKS) homomorphic scheme protects deep learning models from adversary attacks and prevents sensitive raw
data sharing. The aggregator uses a secure federated averaging method that averages encrypted parameters to
provide a global model protecting users’ privacy. The system achieved an accuracy rate of 84.49% when evaluated
using the MIT-BIH arrhythmia dataset. Furthermore, other comprehensive performance metrics were computed to
gain deeper insights, including a precision of 72.84%, recall of 51.88%, and an F1-score of 55.13%, reflecting a
better understanding of the adopted framework. Our findings indicate that employing the CKKS encryption scheme
in a federated environment with transfer cutting-edge technology achieves relatively high accuracy but at the cost
of other performance metrics, which is lower in the encrypted settings when compared to the plain one, an
acceptable trade-off to ensure data privacy through encryption with achieving an optimal model performance.

Design and Development of IoT-Enabled Power Management System for Smart Home with Enhanced Security Feature

indu bala — 2024-08-16

This article presents a proposed architecture for a smart and secure home automation system that incorporates
advanced Internet of Things (IoT) capabilities. The system under consideration has been developed with the
objective of maximizing the efficiency of power consumption in domestic electrical appliances. By employing a
continuous monitoring mechanism for the power consumption of individual devices, this system serves the twin
purpose of mitigating the risk of short circuits resulting from excessive electrical loads and enabling users to
monitor the energy consumption of each appliance. Consequently, it facilitates the efficient management of billing
processes. The proposed model utilizes Internet of Things (IoT) technology to integrate video surveillance at the
residential entry, enabling live monitoring of visitors. The fundamental elements of this novel design comprise an
Arduino board and a collection of sensors encompassing fire, temperature, humidity, and other variables. The
sensors establish a connection with the system via the Thingspeak application, facilitating users' ability to retrieve
real-time power usage data from a range of devices. In the event that a device exceeds a pre-established power
threshold, the user will be promptly notified through the delivery of timely warnings. These alerts will be presented
in the form of graphical representations, which will effectively illustrate the patterns in power consumption.
Moreover, the system's scalability enables the automation of many devices, making it a highly attractive solution
for applications beyond residential settings. This includes industrial automation, notably within the framework of
Industry 4.0.

Robust Image Watermarking Based on Lifting Wavelet Transfrom-Hessenberg Decomposition for Copyright Protection

Agung Sudrajat — 2024-08-25

The modern internet technology enables unauthorized individuals to alter the content of digital
images. This work introduces a method for enhancing the robustness of an embedded watermark in images using
Lifting Wavelet Transform (LWT) and Hessenberg Decomposition. This study utilizes Hessenberg Decomposition
(HD) to insert a watermark image into the LL-sub band of the LWT transform. Singular Value Decomposition
(SVD) is employed to convert the H value of HD. Subsequently, the watermark image is put into the singular
value. The suggested system underwent testing against a range of image processing assaults, including
compression, filters, and noise additions. The testing was conducted using multiple watermark sizes, specifically
256×256, 128×128, and 64×64 pixels. The experimental results demonstrate exceptional imperceptibility, with an
average PSNR value of 39.5640 dB and a SSIM value of 0.9993. The results demonstrate a high level of resilience,
as indicated by the NC value of 0.96390.

Hybrid Honey Badger Algorithm with Artificial Neural Network (HBA-ANN) for Website Phishing Detection

MUHAMMAD ARIF MOHAMAD — 2024-08-18

Phishing is a sort of cyberattack that refers to the practice of fabricating fake websites that imitate
authentic websites in order to trick users into disclosing private information. Identifying these fake sites is
challenging due to their deceptive nature as they often mimic legitimate websites, making it difficult for users to
distinguish between the real and fake ones. Artificial Neural Network (ANN) is one popular method for website
phishing detection. ANN is capable of detecting phishing websites by identifying patterns and characteristics
connected to phishing websites through a network training phase. Technically, in the network training phase of
ANN, neurons on the network must be passed over. There are multiple techniques in training the network, one of
which is training with metaheuristic algorithms. Metaheuristic algorithms that aim to develop more effective
hybrid algorithms by combining the good and successful aspects of more than one algorithm are algorithms
inspired by nature. Therefore, this study proposed a hybrid Honey Badger Algorithm with Artificial Neural
Network (HBA-ANN) classification model. HBA as metahueristic algorithm is used to optimize the network
training process of ANN to improve their performances. Three main steps made up the proposed HBA-ANN
classification model: setting up the experiment, optimizing HBA for network training, and network testing. Lastly,
the performance of the proposed HBA-ANN classification model is assessed in terms of recall, precision, F1-score,
accuracy and error rate using the confusion matrix that was generated for analysis. The proposed hybrid HBAANN was found to be effective in identifying the phishing website after conducting an experimental and statistical
analysis.

A Machine Learning Algorithms for Detecting Phishing Websites: A Comparative Study

Mohanned A.Taha — 2024-07-23

Phishing website attacks are a type of cyber-attack in which perpetrators create fraudulent websites
that mimic legitimate platforms, such as online banking or social media, with the intent of tricking unsuspecting
users into divulging sensitive information. This includes passwords, credit card details, usernames, and other
personal data. These phishing websites are designed to look authentic and often employ various techniques, such as
URL spoofing, social engineering, and email or text message phishing, to lure victims into revealing their
confidential information. Web apps are growing increasingly complex and difficult to identify at first glance,
especially when they use encryption and obfuscation techniques. In order to effectively detect and stop phishing web
applications from being uploaded to the server in real-time, machine learning must be developed. In addition to
including analyses for the machine learning algorithms for identifying web application-based assaults, the study
calibrates fresh analyses by executing machine learning algorithms and confirming the findings. The study uses
unique and categorized results from a machine learning dataset. As per the outcomes obtained from experimental
and comparative analyses of the applied classification algorithms, the random forest model demonstrated the highest
accuracy, achieving an impressive rate of 96.89%, followed by the decision tree model at 94.57%, and Extreme
Gradient Boosting (XG).

Formulating an Advanced Security Protocol for Internet of Medical Things based on Blockchain and Fog Computing Technologies

rasha halim razaq — 2024-08-31

The Internet of Medical Things (IoMT) is an evolving field in healthcare that connects medical
devices to the Internet to enable efficient data sharing and health information collection. The IoMT aims to
improve the quality of healthcare, facilitate diagnosis and treatment, and enhance patient safety. Nonetheless, the
IoMT networks are usually exposed to multiple security attacks. Also, recent studies indicate that security
protocols contain flaws in protecting patient data. Thus, data must be protected by innovative security protocols. In
our work, we propose a Medical Security Protocol (MedSecP) to support security in IoMT. The proposed protocol
adopts the Twofish encryption, Naive Bayes (NB), and decision tree (DT) within the private blockchain (PBC) Fog
Computing (FC) to build robust security procedures. The Twofish encryption algorithm is used to provide medical
information concealment. In our proposed protocol, the type of data is first determined, and accurate and
appropriate medical decisions are made based on the collected data using a decision tree algorithm, and then rapid
classification of the patient data is done using the Naive Bayes algorithm. Confidential medical data is then
encrypted using the Twofish algorithm to ensure the confidentiality of this data and prevent unauthorized access.
Finally, this encrypted medical data is stored using blockchain technology. Twofish, NB, and DT are organized to
work harmoniously with the PBC. The latter manages and distributes data peer-to-peer in IoMT. We leverage Fog
Computing to speed up decision-making without resorting to the remote cloud. We analyzed our protocol in terms
of security and performance. Our results indicate that MedSecP provides reliable security against attacks as the
protocol demonstrated an average security attack response rate of 97.20%, demonstrating its resistance to external
threats by keeping the encrypted medical data, classified and achieving appropriate medical decisions. In terms of
performance, MedSecP has demonstrated an average security response time of around 50ms, providing fast and
efficient performance. In MedSecP, the highest value for encryption is 0.000015 ms, and decryption is 0.000017
ms when applying the Twofish algorithm which is considered extremely suitable for implementing health systems
operations compared to existing encryption algorithms. Consequently, MedSecP provides lightweight operations in
support of complex security measures that qualify it to support healthcare institutions.

Enhancing Efficiency and Fine-Grained Control in Redactable Blockchains with EPBCHF

shams abdali — 2024-07-06

Blockchain technology has presented a promising decentralized paradigm to preclude trusted third
parties' dominancy. It is a transparent and distributed ledger initially designed for digital cryptocurrencies while
currently extended to serve various industries. However, Blockchain immutability presents challenges, as it can be
misused for storing illicit content, violating privacy regulations, and limiting data management flexibility. Policy
Based Chameleon Hash Function (PBCH) has transformed blockchain rewriting contents concept via permitting
modifiers to amend certain transaction since they possessed fundamental privileges satisfying certain access policy.
However, PBCHF suffers from efficiency issues due to its reliance on Chameleon Hash ephemeral Trapdoor (CHET)
and Attribute-Based Encryption (ABE), significantly impacting overall efficiency. We propose the Efficient PolicyBased Chameleon (EPBCHF) construction by replacing CHET with Chameleon-Hashes by Dual Long-Term
Trapdoors (CHDLTT) to address these challenges. Additionally, we introduce an enhanced encryption scheme
resilient against chosen-ciphertext attacks (CCA) without compromising overall efficiency. Modelling EPBCHF
proves practical instantiation accompanied by rigorous security proofs. Our construction provides a fine-grained
redactable blockchain in comparison to the currently proposed solutions. The evaluated results confirm that the
proposed EPBCHF is scalable and efficient due to having the ability to handle unlimited transaction volumes
additionally, data is efficiently processed without further overhead meanwhile data size consistency reflects a robust
memory management due to predicted memory size, network bandwidth and storage requirement for future growth
thereby, EPBCHF is proven to be reliable and scalable.

Adaptive Heterogeneity Index Cloudlet Scheduler for Variable Workload and Virtual Machine Configuration

Gritto D — 2024-08-29

In any service-based computing environment, performance pertains to the effectiveness of a system or application in managing user tasks. The key performance assessment metrics include makespan, responsiveness, speed, throughput, resource utilization, etc. In any distributed landscape, like cloud computing, optimal performance relies on resource management techniques such as scheduling, load balancing, etc. Cloud environments often exhibit varying levels of heterogeneity arising from the diverse characteristics of cloudlets and virtual machines. This research paper focuses on the impact of this heterogeneity and proposes two scheduling algorithms to address it effectively: the Variance Managed Heuristic Scheduler (VMHS) and the Adaptive Heterogeneity Index Cloudlet Scheduler (AHICS). AHICS aims to minimize makespan, virtual machine underutilization, the degree of load imbalance, and the deviation of completion time among virtual machines. AHICS functions as the main scheduler, whereas VMHS and MaxMin act as sub-schedulers in this proposed work. AHICS is designed to be flexible and adjust its scheduling strategy based on the level of heterogeneity within the cloud environment. AHICS utilizes the VMHS scheduler in scenarios with a low heterogeneity index or the MaxMin scheduler when the cloudlets and the virtual machines characteristics are highly diverse or heterogeneous. This multi-objective AHICS scheduling algorithm harnesses the strengths of both schedulers as a hybrid algorithm. Implemented using the CloudSim 3.0.3 simulator, experimental results demonstrate that AHICS outperforms other heuristic scheduling algorithms, including MinMin, TASA, HAMM, PTFR, and RSSM, in terms of makespan, virtual machine utilization ratio, the degree of load imbalance, and the deviation of completion time among the virtual machines in both low and high heterogeneity levels

Taxonomy, Open Challenges, Motivations, and Recommendations in Driver Behavior Recognition: A Systematic Review

ziadoon al-qaysi — 2024-07-31

Driver behavior has a major role in many of the unpleasant things that happen when driving, such as
crashes or accidents, heavy traffic, abrupt braking, and acceleration and deceleration. Numerous investigations
have been carried out to look into the variables influencing driving behavior. To offer a thorough analysis and
classify these findings according to a logical classification, further research is required. The goal of this systematic
review is to enhance knowledge about the variables influencing driving behavior. A taxonomy on the subject of
driver behavior in various ITS domains and their categories was also produced by this work.
A systematic review of the literature was performed in accordance with PRISMA (Preferred Reporting Items for
Systematic Reviews and Meta-Analyses) guidelines to gain insight into driver behaviour recognition.
Specifically, IEEE Explore, ScienceDirect and Springer databases were searched to identify any relevant articles
with a focus on "driver behavior," "driver style," "driver pattern," "driver simulator," and "visual attention,” from
2008 to 2021 (15 April).
Several filtering and scanning procedures were performed on all 606 retrieved articles in compliance with the
exclusion/inclusion criteria; nonetheless, only 50 articles met the requirements. The criteria-compliant references
were examined and evaluated. Furthermore, every piece that was included was classified using a taxonomy. The
four categories created by the taxonomy are review, experiment, framework, and other study types. To illustrate
the main gaps in the literature regarding the identification of driving behavior, a discussion and analysis were
presented. This thorough analysis has highlighted the issues and reasons while also pointing forth fresh research
directions.

Various Closed-Form Solitonic Wave Solutions of Conformable Higher-Dimensional Fokas Model in Fluids and Plasma Physics

Israa Ibrahim — 2024-08-06

This work focuses on finding closed-form analytic solutions of a higher-dimensional fractional model,
in conformable sense, known by the (4+1)-dimensional Fokas equation. Fractional partial differential equations
(FPDEs) and systems can describe heritable real-world occurrences. However, solving such models can be difficult,
especially for nonlinear problems. The homogeneous balancing method (HBM) is investigated and extended to
handle the (4+1)-dimensional Fokas equation with Kerr law nonlinearity. The HBM has the ability to solve linear
and nonlinear fractional problems, incorporating the concepts of some fractional calculus principles, including
fractional derivative techniques. It's important to note that there isn't a single and universally applicable method to
solve such equations due to their complexity. The specific form of the equation and the initial or boundary conditions
influence the solution method chosen. The results obtained from the extended HBM are compared to those in the
literature to prove the strategy's efficacy. This paper proposes expanding the HB technique with result analysis to
solve nonlinear FPDEs, demonstrating its feasibility and efficiency

Fuzzy Weighted Zero Inconsistency Method (FWZIC) for Multi-Criteria Decision-Making Weighting Criteria: A Systematic Literature Review

ziadoon al-qaysi — 2024-08-13

Researchers in various fields have consistently used Multi-Criteria Decision-Making (MCDM)
methods for discipline improvement, employing both standard and novel approaches. Selecting a weighting
mechanism for evaluation criteria is critical in MCDM problems. Recognizing the importance of remaining up to
date on such developments in methodology, this study aims to review several innovative methods integrated with
Fuzzy Weighted Zero Inconsistency (FWZIC). Relying on the papers taken from the significant databases: IEEE
Xplore (IEEE), ScienceDirect (SD), Scopus, and PubMed from 23 August 2023 to 30 October 2023, where each
method was read and analyzed for its characteristics and steps. These indexes were deemaed extensive and
dependable enough to encompass the scope of our literature review. A total number of articles, n = 26 were chosen
based on the criteria for inclusion and exclusion have been selected for this systematic review. By using
bibliometric and content analysis, this study examined the developing ways with (FWZIC), as well as study
components (sources, authors, different countries, and affiliations), areas of application, case studies, fuzzy
implementations, hybrid studies (use of other weighting methods), and application tools for these methods. The
results of this literature systematic review (LSR) provide an accurate depiction of each new development related to
the weighting method and its utilization, such as: 1- Extracting the types of development employed in the FWZIC
approach based on Fuzzy Set. 2- Extracting types of aggregation operators. 3- Analyzing integration methods with
FWZIC (hybridized with other methods), and 4- Case study types showing how MCDM approaches may help
decision-makers in a variety of decisions. Also, a set of recommendations has been presented to the researchers for
the development of new method types, as a new direction for future work. This will provide academics and
practitioners in the field of MCDM with valuable insights and significant expertise for data analysis and decisionmaking.

Arabic Chatbots Challenges and Solutions: A Systematic Literature Review

soufiayn ouali — 2024-06-30

Since the beginning of Natural Language Processing, researchers have been highly interested in the idea of equipping machines with the ability to think, understand, and communicate in a human-like manner. However, despite the significant progress in this field, particularly in English, Arabic research remains in its early stages of development. This study presents a Systematic Literature Review on challenges faced in Arabic chatbot development and the proposed solutions. Utilizing the search terms "ARABIC," "CHATBOT," "CHALLENGES," and "SOLUTION," (including synonyms) we systematically surveyed studies published between 2000 and 2023 from Scopus, Science Direct, Web of Science, PubMed, SpringerLink, IEEE Xplore, ACM, Ebesco, and ICI. Moreover, besides Google Scholar and ResearchGate, we employed a manual snowballing technique to discover supplementary relevant research by examining the references of all chosen primary studies. The included studies were assessed for eligibility based on the quality assessment checklist we developed. Out of 3,891 studies, only 64 were deemed eligible. Many challenges were identified, including the scarcity of well-structured datasets. To overcome this (n=35) studies manually collected and preprocessed data. Additionally, Arabic language complexity led (n=53) researchers to adopt pre-scripted rules approaches, followed by generative approaches (n=10), and hybrid approach (n=1). Furthermore, (n=27) studies employed human-based evaluation metrics to assess the chatbot performance. while, (n=11) studies haven’t used any metrics. Based on conducted research, a critical research priority is providing Arabic with high-quality resources, such as an Arabic dataset that includes dialectal variations and incorporates empathy, lexicon corpora, and also a word normalization library. These resources will enable the chatbot to interact more naturally and humanely. Additionally, hybrid approaches have shown promising results, particularly in low-resource languages, such as Arabic. Therefore, more focus should be dedicated into implementing hybrid approaches in chatbot development. Furthermore, evaluating the chatbot performance is still an open domain for further research and contribution, highlighting the need for innovative standardized evaluation methods.

The New Strange Generalized Rayleigh Family: Characteristics and Applications to COVID-19 Data

Alaa A. Khalaf — 2024-06-10

In this paper, we introduce a novel family of continuous distributions known as the Odd Generalized Rayleigh-G Family. Within this family, we present a special sub-model known as the odd Generalized Rayleigh Inverse Weibull (OGRIW) distribution. The OGRIW distribution is derived by combining the T-X family and the Generalized Rayleigh distribution. We provide a comprehensive expansion of the (PDF) and (CDF) for the OGRIW distribution. Additionally, we investigate several mathematical properties of the OGRIW distribution, including moments, moment-generating function, incomplete moments, quantile function, order statistics and Rényi entropy. To estimate the model parameters, we employ the maximum likelihood method, aiming to identify the parameter values that maximise the likelihood of the observed data.

Finally, we apply the proposed OGRIW distribution to two real COVID-19 datasets from Mexico and Canada. The results of these applications demonstrate that the new distribution exhibits remarkable flexibility and outperforms other comparative distributions in terms of accurately modelling the COVID-19 data.

Approach for Detecting Arabic Fake News using Deep Learning

khalid shaker Alhity — 2024-08-29

: Fake news has spread more widely over the past few years. The development of social media and
internet websites has fueled the spread of fake news, causing it to mix with accurate information. The majority of
studies on Fake News Detection FND were in English, but recent attention has been focused on Arabic. However,
there aren't many studies on Arabic fake news detection. In this work, a new Arabic fake news detection approach
has been proposed using Arabic dataset publically available and a translated English fake news dataset into Arabic.
A new model Text-CNNs based on 1D Convolution Neural Networks CNNs has been used for classification real
and fake news. Extensive experimental results on the Arabic fake news dataset show that our proposed approach
provided high detection accuracy about (99.67%), Precision (99.45), Recall (99.65) and F1-score (99.50).

A Review of the Integration Between Geospatial Artificial Intelligence and Remote Sensing

abeer alshiha — 2024-07-08

: Experts have extensively explored the advantages and applications of modern artificial intelligence
(AI) algorithms across various domains. Geomatics data processing is no exception, as AI offers significant
opportunities in this field. However, understanding how AI can be customized or developed to meet the unique
requirements of geomatics data is crucial. Integrating AI techniques into geomatics has given rise to Geospatial
Artificial Intelligence (GeoAI), a novel approach to uncovering geographic information. Nevertheless, there
remains a shortage of comprehensive research on the specific applications of AI in geospatial contexts.
Consequently, this study aims to establish AI-based methodologies for the analysis and interpretation of complex
geomatics data, bridging existing gaps, and elucidating the connections between AI principles and geomatics data.
This paper delves into the innovations and tools employed for data acquisition in geomatics, focusing on RGB
images, thermal images, 3D point clouds, GPS coordinates, and hyperspectral/multispectral images. Subsequently,
we elucidate how AI techniques have successfully extracted valuable insights from geomatics data. Furthermore,
we present various practical scenarios where AI has been deployed and the specific methodologies employed for
each case. Through this exploration, we aim to highlight the immense potential of AI in geomatics and stimulate
future research endeavors.

LEARNERS' EMOTIONS ESTIMATION USING VIDEO PROCESSING TECHNIQUES FOR OPTIMUM E-LEARNING EXPERIENCE

Mohammed Subhi — 2024-08-16

Learning management systems (LMSs) have integrated multiple technologies to enhance the elearning experience. One such technology is the emotional recognition system (ERS), which provides tutors with
data on learners' emotions, including anger, sadness, happiness, and more. ERS utilizes various data sources like
facial expressions, body activities, and brain signals to recognize emotions. This paper provides an overview of the
ERS structure and discusses the state-of-the-art technologies in this field. The results indicate that deep learning
based ERS using VGG19 for feature extraction over the FER2013 dataset is reliable with a recognition accuracy of
87% using Random Forest Algorithm.

Emotion Recognition Using Various Measures and Computational Methods: Review

Abir Hamrouni — 2024-07-31

Emotion recognition has garnered significant attention as a burgeoning research domain, owing to
its potential applications across diverse fields such as human-computer interaction, affective gaming, marketing,
and human-robot interaction. Accurately interpreting and appropriately responding to human emotions remains a
critical challenge in the development of systems. This obstacle necessitates a thorough understanding of emotions
to enhance user experiences within such systems. This paper conducts a comprehensive review focusing on
advancements in emotion recognition techniques, with an emphasis on leveraging a variety of sensors and
computational methods. Our study findings highlight the significant improvement to emotion recognition accuracy
when multiple measures and computational methods, rather than a single modality, is used. This article contributes
to the field by thoroughly reviewing and comparing diverse measures and computational methods for emotion
recognition. The study highlights the pivotal role of employing multiple modalities and advanced machine learning
algorithms to achieve superior accuracy and reliability in emotion recognition. Furthermore, this research identifies
potential avenues for further investigation and development, such as integrating multimodal data and exploring
novel features and fusion techniques. The insights offered in this study provide valuable guidance for researchers
and practitioners in the field, facilitating the advancement of technologies that adeptly understand and respond to
human emotions.

Automatic Temperature Control System Using African vultures optimization algorithm

Muntadher Khamees — 2024-08-25

One of the most important tasks in control engineering is tuning a PID controller for maximum
efficiency. However, without a great deal of practice, manual adjustment of PID settings might result in erroneous
results. Using met heuristic algorithms is one method for tweaking the PID controller. These algorithms, which are
inspired by the laws of nature, can effectively find the sweet spot for the PID settings. Therefore, instead of
manually tweaking the PID controller, using met heuristic methods can greatly enhance the system's performance
while decreasing the related expenses. A reliable temperature control system is crucial to the production of a wide
variety of industrial goods. The fact that it uses less energy than rivals makes it a hot commodity. The console is
the most effective way to monitor this procedure. The air conditioning in the room is managed by a whole new
model built on metaheuristic algorithms. The temperature regulation is processed using a PID controller based on
the AVOA algorithm. Overshot, settling, and rising times have all been shortened. along with other methods from
the literature, such the Ziegler-Nichols and PSO Optimizer algorithm. Efficiency was evaluated in terms of percent
overrun, reaction time, settling time, and steady state error. Integral Time Squared Error (ITSE), Integral Time
Absolute Error (ITAE), Integral Absolute Error (IAE), and Integral Squared Error (ISE) are used to compare the
empirical data.

SUMER: A New Family of Lightweight and Traditional Block Ciphers with Multi-Modes

omar dawood — 2024-07-08

With the recent increase in the risks and attacks facing our daily lives and digital environment around us,
the trend towards securing data has become inevitable. Block ciphers play a crucial role in modern crypto-applications
such as secure network storage and signatures and are used to safeguard sensitive information. The present paper
develops a new variant of the symmetric model called SUMER family ciphers with three equivalent modes: lightweight,
conventional (traditional), and extended ciphers. SUMER name belongs to one of the oldest civilizations in
Mesopotamia and stands for Secure Universal Model of Encryption Robust Cipher. The SUMER cipher is based on a
simple and robust symmetric structure and involves solid algebraic theories that completely depend on the Galois Field
GF(28
). SUMER cipher is designed to work with two involutional structures of the Substitution-Permutation Network
(SPN) and Feistel structure. These two involutional structures mean that the same algorithm is used for the encryption
and decryption process, and only the algorithm of the ciphering key is used in reverse order in both structures. The
SUMER lightweight structure is an elegant mode that does not need building an S-Box that requires a large amount of
memory and a number of electronic logical gates as S-Box construction has been canceled and replaced by the on-fly
computation clue, which does not need a reserved memory for building S-Box. SUMER family ciphers also can work in
a traditional mode or as an extended mode with high margin security. This family of ciphers is applicable with multimodes of various utilizations. The proposed ciphers are designed to be byte-oriented, showing good evaluation and
results under several measurement tests for speed, time implementation, and efficiency.

Fuzzy Decision by Opinion Score Method (FDOSM): A Systematic Review

ziadoon al-qaysi — 2024-08-13

Multiple criterion decision-making (MCDM) has been widely utilized in everyday life in various
ways, with countless success stories aiding in the analysis of complicated problems and the provision of an
accurate decision process. To date, MCDM remains the best strategy to provide the finest solutions to solve
complex problems in the field of specialized systems. Even so, several challenges faced MCDM approaches, as
mentioned in academic literature. The most significant challenges are uncertainty and ambiguity. The fuzzy
decision by opinion score method (FDOSM) is one of the most recent MCDM methods. Consequently, the purpose
of this study was to examine and assess publications regarding the various types of developments of FDOSM in
recent years and to collect the essential literature findings. Fundamentally, based on systematic review protocol,
the searching process was conducted across four major databases, including IEEE Xplore (IEEE), ScienceDirect
(SD), Scopus, and PubMed from the date of publishing of FDOSM article, based on particular Query "(MCDM OR
'Multi-criteria decision-making') AND (FDOSM OR 'Fuzzy decision by opinion score method') AND (Fuzzy)
AND ('development method' OR 'developing'). The data collection process started on 23 August 2023 and ended
on 30 October 2023, which included several scientific studies related to development FDOSM across several fuzzy
types. These indexes are considered extensive and dependable enough to have the scope of our literature review.
A final set of articles, n=22, was selected depended on predefined inclusion and exclusion criteria for this study.
A coherent classification for current studies has been formed according to the development of FDOSM based on a
new extension of a fuzzy set. Other aspects, such as the research method followed, the protocol adopted by the
systematic review, and demographic statistics of the literature distribution, were included. Exciting patterns were
observed, which were compiled and analyzed in tabular format according to their importance. The results of this
literature systematic review provide a precise summary of each recent development concerning the FDOSM and its
use, including: 1- Extracting the development types employed in the FDOSM method, 2- Extracting aggregation
operator types, 3- Integration Method with FDOSM (hybrid with other methods), and 4- Case study types showing
how FDOSM approaches may help decision-makers in numerous decisions. Finally, this study highlights research
opportunities and encourages current and future efforts to better comprehend this field.

Mobile Augmented Reality Framework in STEM Education: A Systematic Literature Review

Danakorn Nincarean — 2024-08-11

There is a growing concern with preparing a sufficient number of highly qualified professional in
areas of Science, Technology, Engineering and Mathematics (STEM). However, the number of students to reach
proficiency and pursue in STEM majors seems to be decreasing. Augmented Reality (AR) is one of the evolving
technologies that have potential and effect on learning, especially in STEM education. The aim of this paper is to
review and analyze existing works that are related to AR in STEM Education. In this study, the common design
elements, components or features for designing educational AR framework in STEM education were analyzed.
This paper determined a literature review which included 67 studies from scientific journals and papers. The
findings indicated that there were many components which played an important role to construct a mobile AR
framework in STEM education. By reviewing existing research, the study identified key components for designing
effective AR frameworks, including AR systems, multimedia elements, teaching methods, and learning outcomes.
The findings suggested that AR could enhance student motivation, understanding, and development of practical
skills in STEM subjects. With these reviewed, understanding on how to develop the educational AR apps in STEM
education could be enhanced.

Enhanced Cancer Subclassification Using Multi-Omics Clustering and Quantum Cat Swarm Optimization

Mazin Mohammed — 2024-08-13

Integrating multiple omics data can significantly improve the accuracy of cancer subclassification, a
challenging task due to the high dimensionality and limited sample sizes. The integration of these data sets can
enhance model performance. This study addresses these challenges by employing Quantum Cat Swarm
Optimization (QCSO) for feature selection, along with K-means clustering and Support Vector Machine (SVM) for
classification. Using QCSO, the most significant features were identified, resulting in an increase in accuracy from
81% to 100%. Performance was evaluated using accuracy, F1-score, precision, recall, ROC, and the silhouette
metric, all of which confirmed the effectiveness of the feature selection approach. Additionally, this method
enhances the classification of samples while making the models more interpretable, providing better insights into
the molecular mechanisms of cancer. This work contributes to advancing knowledge in the field of cancer research
and biology in general.

The Gompertz Nadarajah-Haghighi (GoNH) Distribution Properties with Application to Real Data

Dalya Dheab Ahmed — 2024-08-25

In this study, we propose a continuous statistical distribution consisting of four parameters based on
the Gompertz family called the Gompertz Nadarajah-Haghighi (GoNH) distribution. Adding parameters to the
basic distribution provides the distribution with flexibility and efficiency in analysing real-world data. The model
that was recently suggested has many mathematical and statistical properties. Explicit formulas for its moments,
moment-generating function, survival function, risk function, characteristic function, quantile function, expansion
of pdf, and ordered statistics are only a few of the many mathematical and statistical features of the recently
proposed model. The maximum likelihood estimates (MLE) method was used to estimate the model’s parameters.
We conducted various simulation experiments to thoroughly evaluate the small sample of MLEs. The research
examined the estimators’ bias and mean square error, yielding positive outcomes. The study’s results demonstrated
that the GoNH distribution fit better than other distributions in two real-world data applications.

Comparison of non-linear time series models (Beta-t-EGARCH and NARMAX models) with Radial Basis Function Neural Network using Real Data

Hiba Abdullah — 2024-06-10

This paper presents a comparison of three different non-linear time series modelling approaches: NARMAX (Non-linear Autoregressive Moving Average with Exogenous Inputs), Beta-t-EGARCH (Beta t Exponential Generalized Autoregressive Conditional Heteroscedasticity), and Radial Basis Function Neural Networks (RBFNN) applied to weekly stock market index data.

We will explain three types of models and compare their compositions and structures. Then, we will show which model gives better predictions. To study series data, the comparison involved analysing the structure of the model and its errors in various time series models and summarising their findings. We divide the data into two parts: training data to structure the time series and testing. The training data tests the model's predictions. Then, we can analyse the model with the errors and the best deterrence predictions. After selecting the NARMAX and Beta-t-EGARCH models, we test them with specific criteria. The best choice is finding the model with the lowest average errors.

For this study, we analysed the weekly average closing of the Aramco 2222 index from 15 December 2019 to 16 July 2023 and made 187 observations.

Soft Computing-Based Generalized Least Deviation Method Algorithm for Modeling and Forecasting COVID-19 using Quasilinear Recurrence Equations

Mostafa Abotaleb — 2024-08-08

This study introduces an advanced algorithm based on the Generalized Least Deviation Method
(GLDM) tailored for the univariate time series analysis of COVID-19 data. At the core of this approach is the
optimization of a loss function, strategically designed to enhance the accuracy of the model’s predictions. The
algorithm leverages second-order terms, crucial for capturing the complexities inherent in time series data. Our
findings reveal that by optimizing the loss function and effectively utilizing second-order model dynamics, there is a
marked improvement in the predictive performance. This advancement leads to a robust and practical forecasting tool,
significantly enhancing the accuracy and reliability of univariate time series forecasts in the context of monitoring
COVID-19 trends.

Distinguishing between Student-Authored and ChatGPTGenerated Texts: A Preliminary Exploration of Human Evaluation Techniques

Md. Saiful Alam — 2024-07-23

The emergence of ChatGPT has opened up numerous possibilities as a supportive tool in the realms
of education and research. However, the potential for students to engage in plagiarism facilitated by ChatGPT poses
a significant challenge for university faculty members, particularly those possessing limited algorithm literacy and
working in low-resourced educational contexts. The study draws upon the initial experiences and reflections of one
of the authors and is accomplished by the collaboration of all the contributing authors. ChatGPT and forum posts
within the learning environment served as the research tools. Employing the Turing Test, seven key human-detection
techniques for deciphering ChatGPT-generated texts have been proposed in this study, including detecting discourse
particles, conversational indicators, the degree of grammatical flawlessness and clarity, formulaic genre structure,
numbered and sub-sectioned body paragraphs, use of transitional words and self-acknowledgment of ChatGPT’s
non-human nature. This study contributes to the emerging body of literature on ChatGPT by enhancing quality
education, reinforcing academic integrity and catalysing further research endeavours in the development towards
human-detection strategies to combat ChatGPT-facilitated plagiarism.

Designing and Implementing an ECG Reader and Heart Disease Classification Device Using Deep Learning (Comprehensive Review)

Ghasaq Saad Jameel — 2024-08-18

Cardiovascular cases remain a significant health care concern, necessitating new development
methods for their prompt and accurate identification where a promising advancements is demonstrated using Deep
Learning (DL) in automating heart anomalies detection, including irregular heartbeats, heart attacks, and abnormal
heart rhythms, through the analysis of electrocardiogram (ECG) signals, which guided this article to explore the
integration of DL methodologies in examining ECG data to enhance the diagnosis of various cardiac ailments and
delves into the development of novel ECG devices designed to facilitate efficient data acquisition while ensuring
patient comfort and accessibility.
The devices used must provide real-time monitoring, seamless integration with DL models, enabling continuous
and personalized heart disease detection for each patient, hence this paper offers an overview of the synergy
between innovative ECG device technology and advanced DL algorithms heralds a transformative era in heart
disease diagnosis, emphasizing patient-centered care, the current research landscape and the challenges ahead, and
the promising possibilities on the horizon, highlighting the potential to revolutionize cardiovascular healthcare
through early, precise, and personalized heart diseases identification and monitoring

Picard and Adomian decomposition methods for a fractional quadratic integral equation via generalized fractional integral

Alan Jalal Abdulqader — 2024-07-06

The primary focus of this paper is to thoroughly examine and analyze a class of a fractional quadratic
integral equation via generalized fractional integral. To achieve this, we introduce an operator that possesses
fixed points corresponding to the solutions of the fractional quadratic integral equation, effectively transforming the
given equation into an equivalent fixed-point problem. By applying the Banach fixed-point theorems, we prove the
uniqueness of solutions to fractional quadratic integral equation. Additionally, The Adomian decomposition method
is used, to solve the resulting fractional quadratic integral equation. This technique rapidly provides convergent
successive approximations of the exact solution to the given fractional quadratic integral equation, therefore, we
investigate the convergence of approximate solutions, using the Adomian decomposition method. Finally, we provide
some examples, to demonstrate our results. Our findings contribute to the current understanding of fractional
quadratic integral equation and their solutions and have the potential to inform future research in this area.

Enhancing Wireless Sensor Networks Features Using Software-Defined Networking Techniques and ACO Algorithms

Ghassan Abed — 2024-08-11

Wireless sensor networks (WSNs) are the most important networks for things such as monitoring
natural phenomena, agriculture, health care, and so on. There are several challenges associated with using WSNs,
the most important of which are energy consumption, expansion issues, or even data routing issues. There are many
techniques and algorithms that can address these challenges and make it easier to use WSNs. These techniques and
algorithms vary depending on the challenge to be overcome. In this study, Software Defined Network (SDN)
technology was used for the purpose of improving WSNs and saving energy for nodes in those networks. The Ant
Colony Optimization (ACO) algorithm, which is an algorithm that follows ants’ foraging method, was also used.
This algorithm was used to find the shortest data path from the starting node to the target node. The results of the
proposed system was a significant improvement in the performance of WSNs. The system saved energy, removed
dead nodes, and found better and shorter paths to reach the target nodes. The proposed approach was versatile and
adaptable, making it suitable for various WSN applications and deployment scenarios. The results of the proposed
system led to energy savings of up to 0.891 mj and only 35 nodes compared to traditional methods, which can have
up to 74 dead nodes.

Tackling the Berth Allocation Problem via Harmony Search Algorithm

Bilal Ahmed — 2024-08-11

Berth Allocation Problem (BAP) is a renowned difficult combinatorial optimization problem that
plays a crucial role in maritime transportation systems. BAP is categorized as non-deterministic polynomial-time
hard (NP-hard) problems, that is very tough to resolve for optimality within an acceptable timeframe. Many
metaheuristic algorithms have been suggested to tackle this problem, and yet, most of these algorithms have some
drawbacks such as they have a weak ability to explore the solution space (they struggle escaping from local
minima) and they face the difficulties to operate on different datasets. Consequently, the need to either enhance the
existing algorithms or utilize a new algorithm is still necessary. Harmony Search Algorithm (HSA) is one of the
recent population-based optimization methods which inspired by modern-nature. HSA has confirmed its ability to
tackle various difficult combinatorial optimization problems like vehicle routing, exam timetabling to name a few.
However, as far as we are concerned, it has never been applied to tackle the BAP problem. The primary objective
of this article is to examine the effectiveness of HSA in solving BAP by identifying suitable values for the
parameters of the HSA and then applying HSA to tackle BAP. Therefore, in this article, the basic HSA is proposed
to tackle the BAP. The suggested HSA is tested on BAP benchmark (I3 dataset) and compared the results with
other latest algorithms found in the literature. The trial outcomes evidenced that the HSA is promising,
competitive, and that it has surpassed some other algorithms that have solved the same dataset, and the results were
very near to the best-known results. Experimental results also prove the suitability and applicability of HSA in
tackling the BAP.

A Novel Deep Learning Approach for Classification of Bird Sound Using Mel Frequency Cepstral Coefficients

Aymen Saad — 2024-08-18

Monitoring animal populations is one important matter to better understand changes in their
population, behavior, and biodiversity. Bird sounds are the main tool to classify bird species acoustically. The
sounds of birds are an indicator for ecologists as it responds to changes in their environment. The recognition
among a variety of bird species to get important features is computationally expensive. With the unbalanced
classes and scarcity of training data, the performance accuracy is degrading. This paper aims to classify species of
birds using lightweight convolutional neural networks (LWCNNs) basis on using a spectrogram image of Brazilian
bird sounds as a dataset. For extracting spectrogram images, Mel Frequency Cepstral Coefficient (MFCC)
algorithm is used. To prove the high performance of the classifier, ten species of birds with 10,000 spectrogram
images are provided to the classifier. Our LWCNN model achieved a training and testing accuracy of 99.68 % and
92.80 % respectively in 10.54 min with 5 epochs.

Solving tri-criteria: total completion time, total late work, and maximum earliness by using exact, and heuristic methods on single machine scheduling problem

Nagham Muosa Neamah — 2024-06-10

The presented study investigated the scheduling regarding jobs on a single machine. Each job will be processed with no interruptions and becomes available for the processing at time 0. The aim is finding a processing order with regard to jobs, minimizing total completion time , total late work , and maximal tardiness which is an NP-hard problem. In the theoretical part of the present work, the mathematical formula for the examined problem will be presented, and a sub-problem of the original problem of minimizing the multi-objective functions is introduced. Also, then the importance regarding the dominance rule (DR) that could be applied to the problem to improve good solutions will be shown. While in the practical part, two exact methods are important; a Branch and Bound algorithm (BAB) and a complete enumeration (CEM) method are applied to solve the three proposed MSP criteria by finding a set of efficient solutions. The experimental results showed that CEM can solve problems for up to jobs. Two approaches of the BAB method were applied: the first approach was BAB without dominance rule (DR), and the BAB method used dominance rules to reduce the number of sequences that need to be considered. Also, this method can solve problems for up to , and the second approach BAB with dominance rule (DR), can solve problems for up to jobs in a reasonable time to find efficient solutions to this problem. In addition, to find good approximate solutions, two heuristic methods for solving the problem are proposed, the first heuristic method can solve up to jobs, while the second heuristic method can solve up to jobs. Practical experiments prove the good performance regarding the two suggested approaches for the original problem. While for a sub-problem the experimental results showed that CEM can solve problems for up to jobs, the BAB without dominance rule (DR) can solve problems for up to , and the second approach BAB with dominance rule (DR), can solve problems for up to jobs in a reasonable time to find efficient solutions to this problem. Finally, the heuristic method can solve up to jobs. Arithmetic results are calculated by coding (programming) algorithms using (MATLAB 2019a)

Prompt Engineering: Unleashing the Power of Large Language Models to Defend Against Social Engineering Attacks

Ahmed I. Nezer — 2024-08-06

Prompt Engineering is an emerging area of study that pertains to the act of conceptualizing, perfecting, and executing prompts that guide an AI model to an intended purpose. The AI model is an LLM, which they are the “hit” of our time and probably the controversial type of AI. They are capable of executing several tasks using natural language processing algorithms. Due to their ease of use and fast development, they are becoming highly dependent. We found that to interact correctly with these models and gain the best performance, several techniques should be taken into consideration. Moreover, there are additional methods or tips to write a good prompt.

Detection of anomalies and Data Drift in a time-series dismissal prediction system

Nataliya Boyko — 2024-07-08

The purpose of the study is to develop a system that automatically processes data based on existing
and newly entered data, especially with the aim of ensuring high data quality by detecting and eliminating
anomalies. The quantile filtering method, Chebyshev’s inequality, Kolmogorov-Smirnov two-sample test, and
others should be noted among the methods used. In the course of the research, the theoretical aspects of the
methods, various principles of detecting anomalies for different types of data were considered and analysed.
Different principles and approaches applied to anomaly detection in different contexts were explored. The results
of the analysis and the selection of optimal methods for detecting anomalies in various types of data are important
for the effective functioning of the automatic data processing system. This will make it possible to achieve
accuracy and reliability in the detection of anomalies and ensure high quality of data used in the machine learning
system.

An Integrative Computational Intelligence for Robust Anomaly Detection in Social Networks

Helina Rajini Suresh — 2024-09-07

Anomaly detection is very important in social networks to keep the truth, security and believability of online communities. This paper presents Adapto Detect which uses a fresh anomaly detection scheme named Pufferfish Optimization Technique (POT) for selecting features and Graph Embedding Autoencoder (GEAE) as an anomaly identifier. POT can pick out vital features from social network data well, it concentrates on attributes necessary for spotting anomalies. At the same time, GEAE is also learning low-dimensional representations of graph nodes. This helps to capture more complicated patterns and relationships within the structure of the graph. These embeddings are used for efficient anomaly detection, which shows deviations from regular social network model behaviour. The performance of AdaptoDetect is superior as shown by its thorough evaluation and comparison with existing methods on different social network datasets and situations. The POT and GEAE interaction in AdaptoDetect allows it to adjust for network alterations and effectively handle anomalies. In general, this system offers a strong answer that can easily identify irregularities within social networks – boosting their security, toughness, and trustworthiness online.

Intelligent Household Load Identification Using Multilevel Random Forest on Smart Meters

Israa Al-Mashhadani — 2024-07-31

A load identification approach for residential intelligent meters using a random forest (RF)
algorithm is employed to guarantee the secure and cost-effective functioning of the electricity grid. In this study,
the load data from a smart meter in a home was pre-processed to remove any gaps, noise, or inconsistencies before
making any predictions by using the random forest method. The power quality (PQ) features, current features, and
Voltage-Current (V-I features), as well as the forecast findings and mathematical tools were used to recognise the
load. Using these tools, the household intelligent meters utilising the random forest algorithm, features, harmonic
characteristics, and instantaneous characteristics were extracted to form the load characteristics, and the objective
function of load identification was generated based on a set of features. The findings of this comparative study
demonstrate that employing this technique can reduce identification errors and boost productivity by a full two
seconds. The proposed approach, based on a random forest technique, improved home power savings rate by
99.2% and the load management efficiency by 98.6%.

Improves Intrusion Detection Performance In Wireless Sensor Networks Through Machine Learning, Enhanced By An Accelerated Deep Learning Model With Advanced Feature Selection

Hadeel M. Saleh — 2024-08-29

Wireless Sensor Networks (WSNs) have been securing a big position in the new aspect of security network attacks, where these suffer from various serious cyber threats that can play with their data integrity and reliability. Due to the key importance of WSN in a wide spectrum range of applications such as environmental monitoring and military field, building reliable, robust and efficient intrusion detection systems (IDS) is necessary. Although traditional machine learning approaches have been intended to detect these threats, they often lack high accuracy due to the complexity and dimensionality of WSN data.

To address these limitations, the study introduces an innovative approach that greatly improves intrusion detection performance in WSNs by combining a high-speed deep learning model with sophisticated feature selection methods. The newly developed system underwent extensive testing using the WSN-DS dataset and applied Gaussian Naive Bayes (GNB) and Stochastic Gradient Descent (SGD) algorithms within the machine learning framework. The outcomes were exceptional, demonstrating a flawless accuracy rate of 100% and representing a significant advancement compared to prior methodologies based solely on traditional machine learning techniques.

The study illustrates how the fusion of deep learning and optimized feature selection effectively addresses the distinctive challenges presented by WSN environments. The results not only present a highly precise and effective method for intrusion detection but also lay the groundwork for further research focused on fortifying the security of sensor networks against progressively intricate cyber threats.