Iraqi Journal for Computer Science and Mathematics

With the increasing prevalence of internet services in our daily lives, there is a growing demand for spectrum, resulting in a shortage of available frequencies. This is primarily due to the recent surge in subscribers seeking faster and more dependable data rates. Researchers globally have suggested various solutions to combat the spectrum shortage, such as technology, modulation schemes, beamforming, intelligent reflective surfaces, and channel coding schemes. With respect to channel coding, multiple codes have been proposed to overcome challenging and uncertain channel conditions. The paper introduces a DWT-incorporated LDPC STBC system as a solution to meet the rigorous demands of 5G deployment and use cases. A comparative analysis of the proposed codes is conducted to assess their suitability for next-generation communication networks, considering SNR and BER. Results indicate that the suggested code performs optimally, achieving a BER of 10-4 at 7dB SNR for Rayleigh Faded wireless channel. Therefore, the suggested code is deemed the most suitable option for the deployment of 5G and beyond systems.

The creation of chatbots, such as Generative Pre-trained Transformer (GPT), is a result of recent developments in natural language processing (NLP). Even though Chat GPT has demonstrated enormous promise in a number of areas, including scientific research, this impact is still developing. This paper attempts to investigate the possibilities, threats, limits, and ethical issues surrounding Chat GPT in scientific research. The assessment of the literature on Chat GPT and scientific research is followed by the presentation of case examples that demonstrate the potential advantages and difficulties of Chat GPT use in scientific research. Finally, we conclude by pointing about the ethical issues that need to be tackled before Chat GPT can be completely utilized in scientific research.

The fuzzy set (FS) is a powerful logic designed to measure uncertain information and plays a crucial
role in solving real-life problems such as decision-making. This article is split into two parts. In the first part, we
highlight the notion of possibility for interval-valued fuzzy soft sets (PI-VFSSs) by collecting all interval fuzzy sets
(IVFSs) with soft sets (SSs) under possibility properties. A basic set theories, some numerical examples, and some
properties for this hybrid model have been created. Then, in order to test the efficiency of the proposed method in
solving some real-life problems, we construct an algorithm using a PIV-FSS. On the other hand, in the second part,
we apply similarity techniques to the proposed model through the analysis of similarity measures of two PIV-FSSs.
After that, a new algorithm based on these measures is applied to the medical diagnosis to find out whether or not
a patient has a respiratory disease. Finally, some theories that show the relationship between PIV-FSS with some
mathematical operations have been proposed.

Wireless sensor networks (WSNs) are targets of intrusion, which seeks to make these networks less capable of performing their duties or even completely eradicate them. The Intrusion Detection System (IDS) is highly important for WSN, since it aids in the identification and detection of harmful attacks that impair the network's regular functionality. In order to strengthen the security of WSN, several machine learning and deep learning approaches are employed in the traditional works. However, its main drawbacks are computational burden, system complexity, poor network performance outcomes, and high false alarms. Therefore, the goal of this study is to develop an intelligent IDS framework for significantly enhancing WSN security through the use of deep learning model. Here, the min-max normalization and data discretization operations are carried out to produce the preprocessed dataset. Then, an Intelligent Prairie Dog Optimization (IPDO) algorithm is used to reduce the dimensionality of features by identifying the best optimal solution with a higher convergence rate. Moreover, a Deep Auto-Neural Network (DANN) based classification method is used to properly forecast the class of data with less false alarms and higher detection rate. For evaluation, a thorough analysis is conducted to evaluate the performance and detection results of the proposed IPDO-DANN model.

This work presents More-SPEED, a novel model for accurately predicting protein activity while minimizing computational demands. Leveraging optimized structures and data preprocessing techniques, More-SPEED achieves high accuracy in protein activity prediction. The model incorporates the data compression three dimension (DC-3D) layer, utilizing the graph mining pattern-fist frequency graph mining (GMP-FFGM) algorithm for efficient preprocessing of complex Deoxyribonucleic acid (DNA) sequence datasets. Additionally, the deterministic structure network using the natural-inspired optimization algorithm called Whale Optimization Algorithm (DSN-WOA) structure optimizes parameters of the Biological dynamic long short term memory (BDLSTM) model, reducing processing time and eliminating manual parameter selection. The BDLSTM layer plays a crucial role in matching codons and predicting protein names, reducing computational complexity without compromising accuracy. The Bi-Rule layer efficiently determines protein activity, especially in disease contexts, providing valuable insights in a shorter time compared to alternative approaches. Evaluation metrics validate the effectiveness of More-SPEED in accurately predicting protein activity, making it a promising solution for advancing protein research.

Electroencephalography (EEG) has been used for quite some time as a diagnostic technique in
neurology. The goal of this publication is to serve as a resource for researchers interested in applying deep learning
methods to EEG data. This paper proposes a unique Hybrid Machine-Deep Learning model that can learn and classify
EEG signals on its own. This method allows the model to classify EEG signals of varied sampling frequencies and
durations automatically. The proposed model used feature extraction methods from artificial design and performed
extensive tests with EEG data collected at varying sample rates to determine how well our suggested model
performed. The results show that the Hybrid Machine-Deep Learning strategy significantly improves performance,
leading to a remarkable 99.97% classification accuracy. Notably, this method performs exceptionally well when
labeling lower-frequency EEG signals (less than 4 Hz). The proposed model has improved consistency and
robustness, as shown by this study.

Except for critical basic implementations seen in historical records, the sheer quantity of similar frameworks, platforms, and apps available may be deceptive. Striking a balance between 3D certified displays and incorporated complicated features continues to be difficult. Geometrically precise point clouds and meshes are the result of cutting-edge methodology (computer vision, reverse engineering, and digital photogrammetry), cutting-edge technology (LiDAR, laser scanners, and unmanned aerial vehicles), and random research approaches. They are a by-product of the procedure. As the precision, size, and complexity of these things rise, the capabilities of present technology to manage them deteriorate. In this paper we propose a method for Development of a low-cost 3d visualization platform, capable of interactively perform tasks. Moreover, we seek to verify the methodological aspect of implementing steps and simplified algorithms in a virtual reality.

Energy efficiency in modern homes has recently become a significant issue due to the emergence of smart home infrastructure. Numerous public structures, such as homes, hospitals, schools, and other institutions, use more energy. To come close to meeting the actual energy demand, it is crucial that we create as much energy as we can. Machine learning has various advantages for improving the effectiveness and efficiency of smart home systems and appliances, including managing and lowering energy use. Additionally, as a key component of the smart home idea, we explore the potential integration of machine learning-based on some algorithm methodologies ways to improve power energy consumption system and control. The models were used to identify patterns for smart home and variations in energy consumption. This study's conclusions were used to analyze case studies and forecast energy consumption. Detection Change (of used and generation) for all appliances, which excessive foresees energy use and stops a rise in usage. Predict Future Energy use by using meteorological data and maximizing the supply of energy to forecast future energy generation and use. Finally, using five machine learning algorithms, including the Linear Regression (LR), Gradient Boosting Regression (GBoostR), Decision Tree Regression (DTR), Stochastic Gradient Descent Regression (SGDR), and Bayesian Ridge Regression (BRR), we can measure the Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Absolute Error (RMAE), and Root Mean Squared Percentage Error (RMSPE), in order to determine how well models.

The COVID-19 virus had easily affected people worldwide through direct contact. Individuals
diagnosed with positive COVID-19 virus may be affected with many symptoms, such as fever, tiredness, dry cough,
difficulty in breathing, sore throat, chest pain, nasal congestion, runny nose, and diarrhea. An individual can also
be diagnosed with COVID-19 even when he does not have any symptoms or be in contact with an infected person.
Data classification was required due to the size of COVID-19 data that will be analyzed for future countermeasures
determination. Some problems in data classification occurred due to unorganized data, such as time consumption,
human error in complexity of symptom features and the diagnosis process data needed expert knowledge. This study
aimed to use the artificial neural network (ANN) approach, which was multilayer perceptron (MLP) to classify the
COVID-19 data by using patient symptom data. The MLP process involved data collection, data normalization, MLP
design, MLP training, testing, and MLP verification. From the experiments, the MLP method was able to obtain
an accuracy rate of 77.10%. In conclusion, the MLP method could classify the COVID-19 data and achieve a high
accuracy rate

Driver fatigue and lethargy are widely acknowledged as significant contributors to the occurrence of vehicular accidents. During the preceding decade, there were a total of 66,000 motor vehicle accidents, resulting in 22,952 fatalities and 79,545 injuries. It is of the utmost importance to implement measures designed to notify drivers in order to prevent accidents and save lives. The purpose of this research was to develop and implement a Driving Drowsiness Detection System (DDDS) in order to increase road safety. The system employs eye and mouth movements as indicators in order to detect fatigue and reduce the likelihood of accidents. Using a high-resolution camera and Deep Cascaded Convolutional Neural Networks (DCCNN) to accomplish accurate drowsiness detection was the objective of this research. This study's methodology entails the analysis of driver behavior, with a particular emphasis on eye movements captured by a high-resolution camera. The facial region was identified using landmarks extracted from the Dlib toolset, and the "Eyes Aspect Ratio" metric was developed to quantify the level of fatigue. Deep Convolutional Neural Network (DCCNN) output triggers an alarm on the dashboard of the graphical user interface (GUI). The study utilized an image with 450x320 pixels and a frame rate of 60 frames per second. In ideal lighting conditions, drowsy detection accuracy exceeded 99.9%, while in less than ideal lighting conditions, it surpassed 99.8%. Compared to previous research, the present study demonstrated a higher level of accuracy in detecting drowsiness.

The Internet of Things (IoT) has transformed our interaction with the world by connecting devices,
sensors, and systems to the Internet, enabling real-time monitoring, control, and automation in various applications
such as smart cities, healthcare, transportation, homes, and grids. However, challenges related to latency, privacy,
and bandwidth have arisen due to the massive influx of data generated by IoT devices and the limitations of
traditional cloud-based architectures. Moreover, network management, interoperability, security, and scalability
issues have emerged due to the rapid growth and heterogeneous nature of IoT devices. To overcome such
problems, researchers proposed a new architecture called Software Defined Networking for Edge Computing in the
Internet of Things (SDN-EC-IoT), which combines Edge Computing for the Internet of Things (EC-IoT) and
Software Defined Internet of Things (SDIoT). Although researchers have studied EC-IoT and SDIoT as individual
architectures, they have not yet addressed the combination of both, creating a significant gap in our understanding
of SDN-EC-IoT. This paper aims to fill this gap by presenting a comprehensive review of how the SDN-EC-IoT
paradigm can solve IoT challenges. To achieve this goal, this study conducted a literature review covering 74
articles published between 2019 and 2023. Finally, this paper identifies future research directions for SDN-EC-IoT,
including the development of interoperability platforms, scalable architectures, low latency and Quality of Service
(QoS) guarantees, efficient handling of big data, enhanced security and privacy, optimized energy consumption,
resource-aware task offloading, and incorporation of machine learning.

Large language models (LLMs) have become prominent tools in various domains, such as natural language processing, machine translation, and the development of creative text. Nevertheless, in order to fully exploit the capabilities of Language Models, it is imperative to establish efficient communication channels between humans and machines. The discipline of engineering involves the creation of well-constructed and informative prompts, which act as a crucial link between human intention and the execution of tasks by machines. The present study examines the concept of rapid engineering, elucidating its underlying concepts, methodologies, and diverse range of practical applications.

In today’s highly competitive business environment, organizations need to collaborate with strategic partners to provide high quality services at lower cost to maintain their effectiveness and efficiency. Hence, the concept of Supply Chain Management (SCM) emerged to effectively coordinate and manage financial, physical, and information activities within organizations and among all participating external parties. Qualitative and quantitative factors must be considered in SCM to ensure sustainable development. The research investigates the metrics and criteria influencing the process of selecting an Internet service provider (ISP) from 2001 to 2022 to assess the extent to which these factors have met the requirements of current decision-makers. It is appropriate to conduct a literature review in light of the changes in the purchasing process and determine its suitability for choosing an ISP. A comprehensive literature review of 51 articles published during the above period was conducted to ensure coverage of the latest ISP selection criteria. The results indicate a decline in the importance of criteria related to cost and prices and an increase in the importance of criteria such as quality, reliability, and performance, as evidenced by their repeated mention in the literature.

The software development process needs specific and studied steps within a reliable plan to achieve the requirements for the success of any project. Software development life cycle (SDLC) methodologies have provided several models that meet the needs of the various proposed projects. These methodologies present various scenarios that can be applied in the process of developing systems to make them more efficient and predictive. The paper aims to illuminate the paramount Software Development Life Cycle (SDLC) methodologies by conducting a comprehensive review of the pros and cons of the various models widely used for software design. Furthermore, the paper discusses fundamental trajectories that are shaping the future landscape of SDLC methodologies. This review included two main types of software development life cycle approaches such as traditional SDLC (heavy-weight) and agile SDLC (light-weight) approaches. The traditional approach included several models such as the Waterfall model, Iterative model, Spiral model, V-Model, and Big Bang Model. Whereas, the agile approach included various models such as the eXtreme Programming (XP) Model, scrum Model, Feature Driven Development (FDD) Model, and Kanban Model. A comprehensive analytical study of all software development life cycle models was achieved and highlighted their most prominent strengths and weaknesses of them. SDLC methodologies wield substantial ramifications across a multitude of sectors, contingent upon several models tailored to individual developmental and research contexts. In culmination, the paper furnishes an all-encompassing perspective on paramount SDLC models, encompassing two principal paradigms: the traditional and the agile approaches. These encompass fundamental sub-models that encapsulate pioneering models poised for application in system development, thus facilitating their refinement more efficiently and predictably.

Securing sensitive data is of utmost importance for organizations and individuals to maintain data confidentiality, integrity, and timely availability. Additionally, the data of organizations and individuals may face increased risks due to attackers gaining unauthorized access, enabling them to misuse the data for illicit purposes. The consequences of such attacks can be severe, leading to significant financial losses and a breakdown of trust between individuals and organizations. Structured Query Language injection (SQL-i) stands out as one of the most prevalent methods employed to illicitly access data, exploiting a vulnerability in the query statement. This vulnerability grants an attacker swift and effortless access to the data. It consequently allows an unauthorized user to tamper with or erase data, or even hinder legitimate access to it. To counteract these attacks, this research aims to build a model using Machine-Learning (ML) techniques that classifies the type of payloads sent by users. This model aims to reduce the time required for payload classification and to scrutinize the false predictions when classifying SQL requests, along with their ramifications on principles of information security (Confidentiality, Integrity and Availability - CIA). The devised model incorporates a dataset containing harmful and benign payloads to train it, employing Logistic Regression (LR) and Singular Value Decomposition (SVD) techniques. The model demonstrated an impressive accuracy of 98.20%, precision of 98.02%, recall of 99.65% and an F1 score of 98.20%. Furthermore, the time taken to classify a payload was a mere 0.0029 seconds. The constructed model excels in accurately categorizing payloads and significantly reducing processing time, owing to the utilization of the LR model in tandem with SVD, which aids in selecting the most pertinent features for training the model.

A precise prediction of student performance is an important aspect within educational institutions to
improve results and provide personalized support of students. However, the predication accuracy of student
performance considers an open issue within education field. Therefore, this paper proposes a developed approach
to identify performance of students using a group modeling. This approach combines the strengths of multiple
algorithms including random forest (RF), decision tree (DT), AdaBoosts, and support vector machine (SVM).
Afterward, the last ensemble estimates as one of the bets logistic regression methods was utilized to create a robust
and reliable predictive model because it considers The experiments were evaluated using the Open University
Learning Analytics Dataset (OULAD) benchmark dataset. The OULAD dataset considers a comprehensive dataset
containing various characteristics related to the student’s activities thereby five cases based on the utilized dataset
were investigated. The experiment results showed that the proposed ensemble model presented its ability with
accurate results to classify student performance by achieving 95% of accuracy rate. As a result, the proposed model
exceeded the accuracy of individual basic models by using the strengths of various algorithms to improve the
generalization by reducing the potential weaknesses of individual models. Consequently, the education institutes
can easily identify students who may need additional support and interventions to improve their academic
performance.

The particle swarm optimization algorithm (PSO) was applied to penalize the Cox model for predicting long-term outcomes of breast cancer patients. This study makes use of data on 198 patients' breast cancer survival, including their age, estrogen receptor status, tumor size, and grade, as well as the expression levels of 76 genes. The objective was to identify a subset of features that could accurately predict patient survival while controlling for overfitting and model complexity. PSO was used to search for optimal model parameters. The algorithm was designed to minimize a penalized partial likelihood function, which balances the tradeoff between an accurate model and model complexity. The values of the objective function were compared with other feature selection techniques, including the Least Absolute Shrinkage and Selection Operator (LASSO) and Elastic regression, and were found to perform better in relation to predictive accuracy and feature selection. The study results showed that PSO-based Cox models with cross-validation to regularization parameters had higher prediction accuracy than models trained with other feature selection methods. The PSO algorithm identified a subset of features that were consistently selected across multiple iterations, indicating their importance in predicting patient survival. Overall, the study demonstrates the potential of PSO-based feature selection in improving the accuracy and interpretability of Cox regression models for predicting long-term outcomes in breast cancer patients.

This work presents a novel method for securing federated learning in healthcare applications, focusing on skin cancer classification. The suggested solution detects and mitigates data poisoning attacks using deep learning and CNN architecture, specifically VGG16. In a federated learning architecture with ten healthcare institutions, the approach ensures collaborative model training while protecting sensitive medical data. Data is meticulously prepared and preprocessed using the Skin Cancer MNIST: HAM10000 dataset. The federated learning approach uses VGG16's powerful feature extraction to classify skin cancer. A robust strategy for spotting data poisoning threats in federated learning is presented in the study. Outlier detection techniques and strict criteria flag and
evaluate problematic model modifications. Performance evaluation proves the model's accuracy, privacy, and data
poisoning resilience. This research presents federated learning-based skin cancer categorization for healthcare
applications that is secure and accurate. The suggested approach improves healthcare diagnostics and emphasizes
data security and privacy in federated learning settings by tackling data poisoning attacks.