https://csmj.uomosul.edu.iq/index.php/csmj en-US [email protected] (Associate Professor Dr. Mohammed Zaki Hasan) [email protected] (Dr.Zeyad Abd-Algfoor Hasan) Thu, 26 Jun 2025 00:00:00 +0000 OJS 3.3.0.7 http://blogs.law.harvard.edu/tech/rss 60 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49517 <![CDATA[In this research, the reducible method or what is called the transformation method was generalized to the Stratonovich formula used to solve stochastic differential equations (SDEs), and the general formulas for the solutions and their theories were reached and the conditions necessary for reducing the stochastic differential equation were clarified by generalizing this method from the Ito integration formula to the Stratonovich integration formula and the transformation method between them, and these two integration formulas (Ito formula and Stratonovich formula) were applied to a group of diverse examples and the solutions were obtained and drawn (by the MATLAB program) and the results of the solutions for both methods were compared.]]> Ali mahmood alojedi, Abdulghafoor J. Salim Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49517 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49518 <![CDATA[In this research, the application of deep learning methods for the classification of human nail diseases using image analysis is investigated. The aim was to establish a non-invasive, automatic diagnosis tool for different nail conditions, utilizing deep convolutional neural networks (CNNs) for feature extraction. A total of 500 images of nails, divided into seven classes of diseases, were employed for training and testing. Feature extraction was performed using VGG16, ResNet50, and EfficientNetB0, and three machine learning classifiers, AdaBoost, LightGBM, and a Meta Classifier, were applied. The multi-classifier data classifier, the Meta Classifier, did better with 98.0% accuracy, 98.2% precision, 97.9% recall, and 98.0% F1 score when used in conjunction with EfficientNetB0. The study validates the efficacy of AI image diagnostics in non-invasive disease diagnosis, delivering a cost-effective and trustworthy method for early diagnosis, particularly in low-resource areas. The study verifies the accuracy of deep learning models, especially EfficientNetB0, for medical image examination, but extensive clinical validation and dataset acquisition are essential]]> Abeer Mohamad Alshiha, Wai Lok Woo Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49518 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49519 <![CDATA[In [1] , we introduce a new family of the Logistic map, namely the cubic logistic maps L= {L_ (x)=x^2 (1-x):>0,xR}. In this work we study the complex dynamics of this family i.e. L= {L_ (z)=z^2 (1-z): >0,zC}. That is we study the Fatou and Julia sets of this maps. In fact we give a whole description for these sets These two new types of logistics maps can address some of life's problems as shown in the introduction.We prove for any R, L_L is preservers R, critically finite, maps the negative x-axis into positive real line and has no any complex periodic point. Fatou set of these maps has no Siegel disk, Baker domain, and has no Wandering domain so they consist of parabolic domains and basins of attraction. Finally, we use escaping algorithm to construct the Fatou and Julia sets of our maps for various values of .]]> baraa salim ahmed, Salma Muslih M. Farris Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49519 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49522 <![CDATA[Parallel processing is essential in machine learning to meet the computational requirements resulting from the complexity of algorithms and the size of the dataset, by taking advantage of the computational resources of parallel processing that can distribute computational operations across multiple processors. Which contributes to significant improvements in performance and time efficiency. This research demonstrated the impact of parallel processing on the performance and time efficiency of machine learning for pixel-based image classification techniques. The methodology includes pre-processing the Oxford IIIT Pet dataset, from which 4 cat images were selected. The performance of two supervised machine learning classifiers, decision tree, and random forest (10, 100, 500, and 1000 trees) were compared and implemented in two ways with and without parallel processing. The data is split in two ways: the first is by splitting the data by 70% for training data and 30% for testing data and the second is by cross-validation by splitting the data into four folds. The research aims to compare the accuracy and timely scales of machine learning models with and without parallel processing. The results showed a strong predictive power of the algorithms with an accuracy of 97.5%, while the training times were significantly reduced in parallel from 88.83 to 15.88 seconds for the RF100 model for image no. 2. This reflects the effectiveness of parallel processing in improving the performance of machine-learning models for pixel-based image classification. The proposed system was programmed using MATLAB 2021 language tools.]]> Aliaa Shaker Mahmoud, Mohammed Chachan Younis Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49522 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49523 <![CDATA[Mobile ad hoc network is a creation of portable devices; they create a network with infrastructure on the fly that are ever changing. As in the previous network structure, each node performs both the action of a router and a host. Also, nodes cannot be fixed in the network; they can join or leave the network, and this increases the flexibility of the connectivity. There are routing protocols which are used for identification of efficient paths between the nodes in the network so as to seek the determination of the best routes between two nodes. This research show that routing is complex in MANETs and hence it demands the fine tuning of numerous routing protocols. We evaluate the effectiveness of these protocols by analyzing two primary metrics: are the average figures of throughput and average end to end delay. Simulation of this protocol was done using NS2 (Network Simulator) 2. 35, we investigate how well routing protocols fare in terms of different aspects including Size of the packets and number of nodes present]]> MOHAMMED JABER ALAWADI Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49523 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49525 <![CDATA[This research introduces and evaluates two enhanced conjugate gradient methods for unconstrained optimization, building upon the DaiLiao conjugacy condition and further refined through the application of Taylor series expansion. These novel methodologies were rigorously compared against the classical Hestenes-Stiefel (HS) method using a diverse suite of benchmark test functions. The numerical results obtained unequivocally demonstrate a significant improvement in computational efficiency achieved by the proposed methods. Notably, our enhanced methods consistently outperformed the HS method across several critical performance metrics, including a reduction in the number of iterations required for convergence, a decrease in the total number of function evaluations, and an overall faster computation time]]> Basim A. Hassan, Alaa Luqman Ibrahim, abd elhamid mehamdia Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49525 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49526 <![CDATA[In this research, a new conjugacy coefficient is derived for conjugate gradient method(C.G) and a new direction was obtained. In theory, this direction achieve sufficiently desent condition by using strong wolfe line search and global convergence is proved. When contrasted with the starnder HS (C.G) technique, the numerical performance of this approach is very remarkable. The Dolan-More performance profile was applied in order to carry out this determination. The amount of time that the central processing unit (CPU) spends, the number of iterations (NOI), and the number of function evaluations (NOF) all play a role in determining this profile. It was determined through the utilization of the Dolan-More profile that this was the case.]]> Hamsa Th. Chilmerane, Mays Basil Jarjis Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49526 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49527 <![CDATA[The Robust Partial Least Square Regression method is used to handle outliers and increase the explanation proportion, but it does not reduce the average of the mean square error. In this article, three methods are proposed to handle the problem of outliers, reduce the average of the mean square error, and increase the explanation proportion of the predictor and dependent variables. The first proposed method (Iteration) depends on identifying outliers by estimating the initial Partial Least Square Regression and then estimating outliers based on the residuals of those values to obtain the lowest mean square error, while the second and third proposed methods depend on a hybrid process between iteration and robust Partial Least Square Regression. The proposed and conventional methods were applied to estimate PLSR models on data Datasets for various ordinary patients in Iraq. The Dataset provides the patients Cell Blood Count test information that can be used to create a Hematology diagnosis/prediction system. Also, this Data was collected in 2022 from Al-Zahraa Al-Ahly Hospital. The proposed iterative method with higher efficiency provided 5 variables' importance in the projection score that explain the changes in HGB levels.]]> taha Hussien ali, Mahammad Mahmoud Bazid Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49527 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49529 <![CDATA[This study aims to enhance the security of Software-Defined Networks (SDN) byemploying deep learning techniques to detect cyber threats and mitigate attacks. A comprehensive data analysis was conducted, beginning with feature identification and dimensionality reduction using the Gain Information method to filter out redundant features, thereby improving model performance. Additionally, Min-Max normalization was applied to standardize feature ranges, and the SMOTE technique was utilized to balance the dataset and reduce the impact of underrepresented classes. The research compares the performance of three primary deep learning modelsCNN, LSTM, and ANNwith a newly proposed method designed to better differentiate between similar attack categories. The results demonstrate that deep learning models can effectively uncover hidden patterns in network traffic and accurately classify security threats, with the LSTM model particularly excelling in capturing temporal dependencies. While CNN and ANN models showed high accuracy in certain scenarios, they struggled to identify classes with fewer samples, necessitating the use of additional balancing techniques. Conversely, the proposed method showed promise in achieving a balance between accuracy and efficiency, suggesting that further refinement in feature engineering and advanced balancing strategies could enhance its performance. Overall, this study underscores the critical role of integrating deep learning with advanced preprocessing techniques in developing more reliable and effective intrusion detection systems for SDN environments.]]> mohammed sami alghaloom Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49529 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49520 <![CDATA[In recent years, technologies have brought about significant changes in education, with digitization becoming a key feature that can enhance the quality and effectiveness of the educational process. Traditional manual attendance and performance assessment systems face several issues, including time wastage, human error, and data loss. As a result, the adoption of technology-based systems that focus on RFID, Arduino, facial recognition, and machine learning technologies has led to increased efficiency and accuracy by automating these processes. Previous studies have explored a variety of approaches and techniques to achieve these goals, such as using RFID for automatic attendance recording, Arduino systems for automated testing, and machine learning algorithms for analyzing academic performance. However, these systems also face challenges, including financial constraints, privacy concerns, and difficulties in scalability. Therefore, this paper aims to provide a comprehensive review of these studies by examining their methodologies, results, and challenges.]]> Sura Hasan Falih, Yaseen Hikmat Ismael Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49520 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49521 <![CDATA[This study explores the structure and properties of strongly Tri-nil clean rings. A ring is defined as strongly Tri-nil clean if every member in a ring can be expressed as the sum of a tripotent element and a nilpotent member, where these components commute. We provide the right singular ideal of a ring which is a nil ideal. We examine a ring with each element in R,^2 is zhou. we also found in these rings the char(R)=48, and every unit of order 4, Finally, we provide if R is a Tri-nil clean ring with 3N(R) if and only if every member of R is a sum of three tripotent and nilpotent that commute.]]> Rana Mahammed Shafik, Nazar H. Shuker Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49521 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49524 <![CDATA[In todays digital age, credit card fraud has become a serious issue, posing financial risks to individuals, businesses, and financial institutions alike. Detecting credit card fraud is crucial to limiting these risks and securing financial systems. This article presents an improved support vector machine (SVM)-based approach that integrates an advanced feature selection method for identifying fraudulent activities. By using a binary genetic algorithm and cross-entropy, our feature selection approach identifies key attributes and evaluates their relevance to the target variable. The SVM classification model then performs the final classification, with its hyperparameters optimized through the particle swarm optimization (PSO) technique. Experimental results on the Credit Card Fraud Detection dataset demonstrate the effectiveness of this method, achieving an impressive accuracy of 99.99%. By combining advanced feature selection with optimization techniques, this approach enhances the accuracy and efficiency of credit card fraud detection, offering a practical solution to combat fraud in financial systems.]]> Mohammed mansooor AL-Hammadi Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49524 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49528 <![CDATA[Video is a way to convey information on the internet which is more popular than the text. Video processing expanded with the using of videos in various fields and highly developing of communication technologies. The process of identifying the video shots is called video segmentation or video shot boundary detection. The aim of Shot boundary detection is to detect the shot boundaries and transitions between successive shots. Video shot boundary detection technique is one of the major research areas in video processing. There are various techniques for different types of videos proposed in this domain. This paper presents various techniques used for Shot boundary detection previously till now and their performance including their advantages, disadvantages, limitations, and future works.]]> Rafal Ali Sameer Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49528 Sun, 01 Jun 2025 00:00:00 +0000 https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49530 <![CDATA[Software bug prediction (SDP) techniques identify bugs in the early stages of the software development life cycle through a series of steps to produce reliable and high-quality software. Deep learning techniques are widely used in SDP, which can produce accurate and exceptional results in different fields.The study aims to systematically review models, techniques, datasets, and performance evaluation metrics to gain a complete understanding of current methodologies related to SDP, and the use of DL in software defect prediction research between 2019 and 2024. A comprehensive review of studies in this area was completed to answer the research questions and summarize the results from the initial investigations. 30 primary studies that passed the systematic review quality assessment of the studies were used. However, the six most common evaluation metrics used in SDP were confusion matrix, Scoar-1F, recall, precision, accuracy, and area under the curve (AUC). The top three DL algorithms used in building SDP models and used in predicting software bugs were convolutional neural network (CNN), long-short-term memory (LSTM), and bidirectional LSTM. We conclude that the application of deep learning in SDP remains a challenge, but it has the potential to achieve better prediction performance. Future research directions focus on improving these models and exploring their applications across diverse programming environments]]> Akhlas Tariq Hasan, Shayma Mustafa Mohi-Aldeen Copyright (c) https://csmj.uomosul.edu.iq/index.php/csmj/article/view/49530 Sun, 01 Jun 2025 00:00:00 +0000