AlKadhum Journal of Science
Not a member yet
50 research outputs found
Sort by
The Connected Classroom: Leveraging EdTech to Enhance Student Engagement
No full textتناولت هذه الدراسة متعددة الأساليب تصورات طلاب الجامعات لدور التكنولوجيا في تعزيز المشاركة والتعاون والاستقلالية. جُمعت البيانات من 120 طالبًا (65 طالبة و55 ميلاً) من ثلاث جامعات (الجامعة المستنصرية، وجامعة ديالى، وجامعة واسط) من خلال المشاركة في الاستبيانات وملاحظات الفصول الدراسية والتغذية الراجعة المفتوحة. أظهرت النتائج أن الطلاب لديهم وجهات نظر إيجابية للغاية، حيث كانت المشاركة (م = 4.12) هي الميزة الأكثر ملاحظة، تليها التعاون (م = 3.98) والاستقلالية (م = 3.85). كشف التحليل الاستدلالي عن اختلافات كبيرة في تصورات التعاون: أبلغت الطالبات عن درجات أعلى من الطلاب (ت (118) = 2.04، ص = 0.04)، وأبلغ طلاب جامعة المستنصرية عن تعاون أعلى بكثير من طلاب جامعة أخرى (ف (2,117) = 3.67، ص = 0.028). لم يتم اكتشاف أي اختلافات كبيرة في المشاركة أو الاستقلالية. كشفت بيانات الرصد عن مستويات عالية من التعاون بين الأقران، بينما سلّطت الملاحظات النوعية الضوء على الفرص الرئيسية (مثل مرونة التعلم) والتحديات (مثل مشاكل التواصل، وفجوات المعرفة الرقمية). وتشير النتائج إلى أنه على الرغم من فعالية التكنولوجيا في إشراك الطلاب، إلا أن قدرتها على تعزيز التعاون تعتمد على عوامل ديموغرافية ومؤسسية. وتؤكد الدراسة على أهمية التصميم التربوي المدروس والدعم المؤسسي القوي للاستفادة من إمكانات التكنولوجيا التعاونية بكفاءة وعدالة.This mixed-methods study examined university students\u27 perceptions of the role of technology in promoting engagement, collaboration, and autonomy. Data were collected from 120 students (65 females and 55 miles) from three universities (Al-Mustansiriya University, University of Diyala, and University of Wasit) through participation in questionnaires, classroom observations, and open-ended feedback.
Results showed that students had highly positive viewpoints, with engagement (M=4.12) being the most strongly observed advantage, followed by collaboration (M=3.98) and autonomy (M=3.85). Inferential analysis revealed significant differences in perceptions of collaboration: female students reported higher scores than male students (t (118) =2.04, p=.04), and students from Al-Mustansiriyah University reported significantly higher collaboration than those from another (F (2,117) =3.67, p=.028). No significant differences were discovered for engagement or autonomy. Observational data revealed high levels of peer collaboration, while qualitative feedback highlighted key opportunities (e.g., learning flexibility) and challenges (e.g., communication issues, digital literacy gaps).
The results indicate that while technology is an effective tool for student engagement, its ability to enhance collaboration depends on demographic and institutional factors. The study underscores the importance of thoughtful pedagogical design and strong institutional support to leverage the potential of collaborative technology efficiently and equitably
Explainable Artificial Intelligence Integrated Ensemble Learning Framework for Diabetes Prediction
No full textAccurate prediction of diabetes based on clinical and demographic indicators is essential, as early prediction of this chronic metabolic disorder plays a critical role in preventing long-term organ complications. However, existing research continues to face significant challenges, including pronounced class imbalance, the scarcity of large and diverse datasets, and limited integration of explainable artificial intelligence. This research compares several ensemble learning methods (Decision Tree, Random Forest, AdaBoost, Gradient Boosting, Histogram based Gradient Boosting, extremely randomized Trees, and XGBoost) on a large imbalanced dataset (87,664 negative vs. 8,482 positive samples). To mitigate imbalance, we evaluate six resampling approaches including (Random Over Sampling, Random Under Sampling, Synthetic Minority Over-sampling, Adaptive Synthetic Sampling, Tomek Links Removal Sampling and (Synthetic Minority Over-sampling with Edited Nearest Neighbors). We assess models using metrics robust to class imbalance (precision, recall, F1, AUC-ROC, and AUC-PR) and calibration measures. The Extra Trees classifier achieved the highest measured accuracy (0.994); with Random Over Sampling for balancing dataset. also, these results were compared with several previous works and number of machine learning algorithms, and the results showed superiority. Explainability is performed at both global and local levels: permutation and SHAP for global feature importance, and (Local Interpretable Model-Agnostic Explanations) force plots for instance-level reasoning. however, we analyzed this result using sensitivity, specificity, PR-AUC and calibration, we report detailed experiments showing how resampling method, hyperparameter tuning, and stratified validation influence performance. Finally, we provide clinical-relevant insights from SHAP analyses and discuss limitations and future directions for deploying interpretable models in screening workflows
Optimized Hybrid CNN-LSTM Framework with Multi-Feature Analysis and SMOTE for Intrusion Detection in SDN
No full textThere is a growing trend toward the use of software-defined networks (SDN), which presents new security challenges requiring advanced intrusion detection systems (IDS). This paper proposes a deep learning-based hybrid system combining convolutional neural networks (CNNs) and long-term short-term memory networks (LSTMs) that can be used for effective intrusion detection in SDN environments. The model uses CNNs to acquire the spatial features of network traffic data and LSTMs to learn temporal patterns, enabling the identification of complex attack patterns. We evaluate our model using an InSDN dataset and test its performance using various feature sets, ranging from 6 to 83 features in our model. Experimental results indicate that our model has a high multi-class classification accuracy of 99.63% when using all 83 features in Group1. Furthermore, we use a Synthetic Minority Over-sampling Technique (SMOTE) to address the issue of class imbalance which considerably enhances detection accuracy of minority attack classes which is reaching 99.76%. It is established that the presented hybrid CNN-LSTM model is powerful and successful solution to SDN security improvement
Audio Encryption and Decryption using the Affine Cipher with the One-Time Pad (OTP)
No full textيهدف هذا البحث إلى تصميم وتنفيذ نظام تشفير هجين لتأمين الإشارات الصوتية، بالاعتماد على دمج خوارزميتين تشفيريتين: التشفير الأفيني (Affin Cipher) ولوحة الاستخدام الواحد (OTP). تُعد خوارزمية التشفير الأفيني نموذجًا بسيطًا وفعالًا للتشفير الكلاسيكي، بينما توفر خوارزمية لوحة الاستخدام الواحد مستوى أمان عاليًا عند استخدام مفتاح عشوائي غير متكرر. طُبّق النظام المقترح على إشارة صوتية رقمية، حيث حُوّلت الإشارة أولًا إلى تمثيل رقمي (8 بت)، ثم رُمّمت باستخدام التشفير الأفيني، ثم طُبّقت تقنية تشفير OTP على المخرجات الناتجة. تضمّن التقييم العملي استخدام مقاييس كمية مثل متوسط الخطأ التربيعي (MSE) ومتوسط الخطأ المطلق (MAE) ونسبة الإشارة إلى الضوضاء (SNR)، بالإضافة إلى التوزيع الإحصائي للإشارات في كل مرحلة من مراحل التشفير باستخدام الهيستوغرام. أظهرت النتائج أن الإشارة المفككة مطابقة تمامًا للإشارة الأصلية (MSE = 0، MAE = 0، SNR = ∞)، وأثبت التحليل الإحصائي للمدرج التكراري فعالية النظام في تشويش أنماط تردد الإشارة، مما يعزز مقاومته للهجمات التحليلية. يوفر النظام المقترح توازنًا بين بساطة التنفيذ وكفاءة التشفير، وهو مناسب لتأمين البيانات الصوتية في تطبيقات الوقت الفعلي والأنظمة الحساسة.This research aims to design and implement a hybrid encryption system to secure audio signals, based on the integration of two cryptographic algorithms: Affine Cipher and One-Time Pad (OTP). The Affine Cipher algorithm is a simple and efficient model of classical encryption, while the One-Time Pad algorithm provides a high level of security when using a random, unrepeated key. The proposed system was applied to a digital audio signal, where the signal was first converted to a numerical representation (8-bit), then encrypted using Affine Cipher, followed by the application of OTP encryption to the resulting outputs. The practical evaluation included the use of quantitative measures such as the average quadratic error (MSE), the average absolute error (MAE), and the signal-to-noise ratio (SNR), in addition to the statistical distribution of signals at each stage of the encryption by histogram. The results showed that the loose signal was exactly identical to the original signal (MSE = 0, MAE = 0, SNR = ∞), and statistical analysis of the histogram proved the effectiveness of the system in blurring the frequency patterns of the signal, which enhances its resistance to analytical attacks. The proposed system provides a balance between simplicity of implementation and efficiency in encryption, and is suitable for securing audio data in real-time applications and sensitive system
Computationally Efficient Hybrid Framework for MNIST Handwritten Digit Recognition
No full textThis research work suggests a combination classification technique for handwritten digit recognition based on MNIST dataset. The design embraces Principal Component Analysis (PCA) to reduce the level of dimensionality, KMeans clustering to get the pattern of the framework, and Random Forest to make the final prediction. By compressing the feature space into a 50 dimensional space and adding in cluster information, we are able to attain an accuracy of about 93.1 percent. The significant contributions of this paper are to prove the PCA-KMeans integration useful to improve classification, offering a computationally efficient solution as alternative to deep learning model, and to prove that feature augmentation using clustering outcomes leads to better discrimination of visually similar digits. The data suggest that the referred to hybrid model is stable and feasible approach to image identification and pattern recognition, which could be applied to more complicated dataset.
 
Multi-Dimensional Deep Learning for Unusual Detection in Security Footage: A 3D CNN, LSTM, and Attention-Based Approach
No full textلا يزال الكشف الآلي عن الأنشطة غير الاعتيادية في فيديوهات المراقبة يُمثل تحديًا بالغ الأهمية نظرًا للكم الهائل من اللقطات وندرة الأحداث الشاذة. تقترح هذه الدراسة إطار عمل مبتكرًا للتعلم العميق يدمج الشبكات العصبية التلافيفية ثلاثية الأبعاد لاستخراج السمات المكانية الزمنية، وشبكات الذاكرة طويلة المدى قصيرة المدى لنمذجة التبعيات الزمنية، وآلية انتباه للتركيز على الأجزاء البارزة. الهدف الرئيسي هو تحقيق تصنيف ثنائي دقيق للغاية لمقاطع الفيديو إلى فئتين "معتادة" و"غير عادية"، مع معالجة اختلال التوازن بين الفئات والتباين البيئي. تم تدريب النموذج وتقييمه على ثلاث مجموعات بيانات واسعة النطاق، وهي UCF-Crime وXD-Violence وCCTVFights، والتي تتضمن فيديوهات مراقبة تغطي شذوذًا في العالم الحقيقي. تُظهر النتائج التجريبية أن الطريقة المقترحة تُحقق دقة إجمالية تبلغ 97.41% في مجموعة بيانات UCF-Crime، و98.11% في مجموعة بيانات XD-Violence، و98.50% في مجموعة بيانات CCTVFights، بالإضافة إلى دقة عالية، ودرجات تذكر، وF1 في مجموعات البيانات الثلاث المستخدمة في عملية التقييم، متفوقةً بذلك على المعايير الحالية. تشير هذه النتائج إلى أن الجمع بين النمذجة المكانية الزمنية وتجميع السياقات القائمة على الانتباه يُمكن أن يُحسّن بشكل كبير من أداء اكتشاف الشذوذ في سيناريوهات المراقبة المعقدة.Automated detection of unusual activities in surveillance videos remains a critical challenge due to the vast volume of footage and the rarity of anomalous events. This study proposes a novel deep learning framework that integrates 3D convolutional neural networks for spatiotemporal feature extraction, Long Short-Term Memory networks for modeling temporal dependencies, and an attention mechanism to focus on salient segments. The primary objective is to achieve highly accurate binary classification of video clips into “usual” and “unusual” categories, while addressing class imbalance and environmental variability. The model is trained and evaluated on three large-scale datasets, UCF‑Crime, XD-Violence, and CCTVFights, which comprise surveillance videos covering real-world anomalies. Experimental results show that the proposed method achieves an overall accuracy of 97.41% on the UCF-Crime dataset, 98.11% on the XD-Violence dataset, and 98.50% on the CCTVFights dataset, in addition to high precision, recall, and F1 scores on the three datasets used in the evaluation process, outperforming existing benchmarks. These findings indicate that combining spatiotemporal modeling and attention-driven context aggregation can significantly enhance anomaly detection performance in complex surveillance scenarios
Quantitative Assessment of Energy Storage Systems For Enhanced Utilization of Renewable Energy in Agricultural Setting : Quantitative Assessment of Energy Storage Systems For Enhanced Utilization of Renewable Energy in Agricultural Setting
No full textتقدم هذه الدراسة تقييمًا كميًا شاملًا لأنظمة تخزين الطاقة (ESS) لتعزيز دمج الطاقة المتجددة في البيئات الزراعية. ونظرًا للطبيعة المتقطعة لطاقتي الشمس والرياح، تُقدم أنظمة تخزين الطاقة (ESS) حلاً استراتيجيًا لضمان موثوقية الطاقة في العمليات الزراعية الحيوية، مثل الري والمعالجة والتخزين البارد. يستخدم البحث نماذج محاكاة ديناميكية، وبيانات ميدانية من مصر والهند والبرازيل، وتحليلًا تقنيًا واقتصاديًا لمقارنة تقنيات تخزين الطاقة بالبطاريات والطاقة الحرارية والطاقة الكهرومائية. ويتم تقييم مؤشرات الأداء الرئيسية - بما في ذلك كفاءة الطاقة، وتكلفة التخزين المستوية (LCOS)، وخفض الكربون، ورضا المستخدمين - عبر مناطق زراعية مناخية متنوعة. تكشف النتائج عن عدم وجود نظام تخزين طاقة واحد مثالي عالميًا؛ فكل تقنية لها مزاياها الخاصة بناءً على المناخ والتكلفة وقابلية التوسع. وتُسلط الدراسة الضوء على أهمية تصميم نظام مُخصص للمناخ، والدعم المُستهدف، ومبادرات بناء القدرات لدعم اعتماد حلول الطاقة المستدامة في الزراعة. وتهدف النتائج إلى توجيه صانعي السياسات والمهندسين والمزارعين نحو أنظمة طاقة زراعية مرنة ومنخفضة الانبعاثات.This study presents a comprehensive quantitative assessment of Energy Storage Systems (ESS) to enhance the integration of renewable energy in agricultural settings. Given the intermittent nature of solar and wind energy, ESS offers a strategic solution to ensure energy reliability for critical farming operations such as irrigation, processing, and cold storage. The research employs dynamic simulation models, field data from Egypt, India, and Brazil, and techno-economic analysis to compare battery, thermal, and hydropower storage technologies. Key performance indicators—including energy efficiency, Levelized Cost of Storage (LCOS), carbon reduction, and user satisfaction—are evaluated across diverse agro-climatic zones. Results reveal that no single ESS is universally optimal; each technology has trade-offs based on climate, cost, and scalability. The study highlights the importance of climate-specific system design, targeted subsidies, and capacity-building initiatives to support the adoption of sustainable energy solutions in agriculture. Findings aim to guide policymakers, engineers, and farmers toward resilient, low-emission agricultural energy systems
A A Comprehensive Review of Intrusion Detection Systems in IoT networks Using ML and DL Techniques: A Comprehensive Review of Intrusion Detection Systems in IoT Networks Using ML and DL Techniques
No full textThe Internet of Things (IoT) is growing at an extremely rapid rate, impacting all aspects of our lives and extending to various fields, including wearable technology, smart sensors, and home appliances. However, the rapid growth is coupled with serious security concerns that render these technologies vulnerable to hacking opportunities and erode user privacy, as well as data protection, especially as cyber-attacks become more complex. Intrusion detection is a crucial aspect for tracking and thwarting such attacks. Machine learning (ML) and deep learning (DL) algorithms have ever-increasing efficiency in automating procedures like these. This study aims to provide researchers with a comprehensive overview of contemporary Intrusion Detection System (IDS) techniques employed in the IoT environment, highlighting strengths and weaknesses. It also gives direction to future research by suggesting that more adaptive, lightweight, and efficient intrusion detection systems can be developed to address the unique constraints of IoT networks
Incipient Fault Protection using Artificial Intelligence Techniques
No full textIn the field of incipient fault protection, various sources can cause failures, such as lightning, switching transients, mechanical imperfections, and chemical breakdown. To guard against these errors, Buchholz relays and pressure relief devices have been utilized. However, in recent years, preventive health measures have gained more attention. One popular approach is the implementation of the Dissolved Gas Analysis (DGA) system, which detects incipient faults by analyzing the gases dissolved in the transformer oil. In this context, the use of artificial neural networks (ANN) and artificial neural networks combined with expert systems (ANNEPS) has shown promise for power transformer protection against incipient faults using DGA. Power transformers, especially large oil-filled ones, are commonly subjected to DGA for identifying and diagnosing early-stage faults. By analyzing the dissolved gases and employing interpretation systems, such as ANNEPS, unexpected failures can be prevented. The objective of this research is to identify internal problems within transformers, and an ANN structure has been specifically developed for this purpose. The ANNEPS approach combines the outputs of ANN and expert systems to ensure rapid and accurate identification of various types of transformer failures. By comparing the results of both computational methods, a reliable assessment can be made, enhancing the effectiveness of incipient fault protection strategies. Overall, the combination of DGA and advanced techniques like ANN and ANNEPS provides a robust approach to detect and prevent incipient faults in power transformers. These methods offer improved accuracy and promptness in identifying transformer failures, ultimately contributing to the reliability and efficiency of power systems.
In the field of incipient fault protection, various sources can cause failures, such as lightning, switching transients, mechanical imperfections, and chemical breakdown. To guard against these errors, Buchholz relays and pressure relief devices have been utilized. However, in recent years, preventive health measures have gained more attention. One popular approach is the implementation of the Dissolved Gas Analysis (DGA) system, which detects incipient faults by analyzing the gases dissolved in the transformer oil. In this context, the use of artificial neural networks (ANN) and artificial neural networks combined with expert systems (ANNEPS) has shown promise for power transformer protection against incipient faults using DGA. Power transformers, especially large oil-filled ones, are commonly subjected to DGA for identifying and diagnosing early-stage faults. By analyzing the dissolved gases and employing interpretation systems, such as (ANNEPS), unexpected failures can be prevented. The objective of this research is to identify internal problems within transformers, and an (ANN) structure has been specifically developed for this purpose. The ANNEPS approach combines the outputs of ANN and expert systems to ensure rapid and accurate identification of various types of transformer failures. By comparing the results of both computational methods, a reliable assessment can be made, enhancing the effectiveness of incipient fault protection strategies. Overall, the combination of (DGA) and advanced techniques like (ANN) and (ANNEPS) provides a robust approach to detect and prevent incipient faults in power transformers. These methods offer improved accuracy and promptness in identifying transformer failures, ultimately contributing to the reliability and efficiency of power systems
Automated Emotion Recognition Using Hybrid CNN-RNN Models on Multimodal Physiological Signals.: Automated Emotion Recognition Using Hybrid CNN-RNN Models on Multimodal Physiological Signals
No full textلقد برز التعرف على المشاعر كأحد أحجار الزاوية للتفاعل بين الإنسان والحاسوب، وبالتالي فتح آفاق جديدة في الرعاية الصحية والتعليم والترفيه. إن القدرة على أتمتة عمليات التعرف على المشاعر باستخدام نماذج الشبكة العصبية التلافيفية الهجينة والشبكة العصبية المتكررة توفر طريقًا واعدًا لفك تشفير الحالات العاطفية المعقدة. تطور الدراسة المقترحة نهجًا لدمج تخطيط القلب واستجابة الجلد الجلفانية وتعبيرات الوجه لأداء التعرف على المشاعر بطريقة دقيقة وفعالة. يجمع هذا الهيكل الهجين بين نقاط قوة CNN في استخراج السمات المكانية وRNNs في نمذجة التبعيات الزمنية، مما يوفر بشكل طبيعي علاجًا للتحديات التي يفرضها استخدام البيانات المتعددة الوسائط. تم إجراء تجارب مكثفة على مجموعات البيانات المرجعية المتاحة للجمهور، ويتفوق النموذج الهجين المقترح على الطرق الأحادية والتقليدية الأخرى من حيث دقة التصنيف الأعلى والمتانة. لا تشير هذه الدراسة إلى إمكانات النماذج الهجينة في تعزيز التعرف على المشاعر فحسب، بل إنها توفر أيضًا إطارًا قابلًا للتطوير وقابلًا للتكيف مع التطبيقات في العالم الحقيقي مثل مراقبة الصحة العقلية وأنظمة التعلم التكيفي. وقد أكدت النتائج كيف يمكن لتقنيات التعلم العميق سد الفجوة بشكل كبير بين التجارب العاطفية الذاتية والتحليلات الحسابية الموضوعية.Emotion recognition has emerged as one of the cornerstones of human-computer interaction, thus opening new frontiers in healthcare, education, and entertainment. The ability to automate emotion recognition processes using hybrid Convolutional Neural Network-Recurrent Neural Network models offers a promising avenue for decoding complex emotional states. The proposed study develops an approach for the integration of electrocardiogram, galvanic skin response, and facial expressions for performing emotion recognition in an accurate and efficient manner. This hybrid architecture combines the strengths of CNNs in spatial feature extraction and RNNs in modeling temporal dependencies, which naturally provides a remedy for challenges inherently brought about by the use of multimodal data. Extensive experiments have been conducted on benchmark datasets publicly available, and the proposed hybrid model outperforms other unmoral and traditional methods in terms of higher classification accuracy and robustness. This study points not only to the potential of hybrid models in advancing emotion recognition but also provides a scalable framework adaptable for real-world applications such as mental health monitoring and adaptive learning systems. The results underlined how deep learning techniques can dramatically bridge the gap between subjective emotional experiences and objective computational analyses.