Pub Date : 2024-07-15DOI: 10.21608/ijt.2024.291347.1052
Ayman Saber, Ahmed Taha, Khalid Abd El Salam
: Arabic handwriting recognition presents unique challenges due to the complexities of Arabic calligraphy and variations in writing styles. Proposing a novel approach to address these challenges by leveraging advanced deep learning techniques. This focus is on Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (Bi-LSTM) networks, which are tailored specifically for recognizing handwritten Arabic text. Utilizing the KHATT dataset for comprehensive training and evaluation, implementing rigorous pre-processing steps to enhance data quality. Central to this methodology is the Res-Net152 architecture for feature extraction, which has proven highly effective. This approach achieved remarkable results, with a character error rate of approximately 2.96% and an accuracy of 97.04% on the testing dataset. These results significantly outperform the previous method, representing a substantial advancement in the field of Arabic handwriting recognition. The study demonstrates the potential of deep learning models in overcoming the unique challenges posed by Arabic script, paving the way for further improvements and applications.
{"title":"A Comprehensive Approach to Arabic Handwriting Recognition: Deep Convolutional Networks and Bidirectional Recurrent Models for Arabic Scripts","authors":"Ayman Saber, Ahmed Taha, Khalid Abd El Salam","doi":"10.21608/ijt.2024.291347.1052","DOIUrl":"https://doi.org/10.21608/ijt.2024.291347.1052","url":null,"abstract":": Arabic handwriting recognition presents unique challenges due to the complexities of Arabic calligraphy and variations in writing styles. Proposing a novel approach to address these challenges by leveraging advanced deep learning techniques. This focus is on Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (Bi-LSTM) networks, which are tailored specifically for recognizing handwritten Arabic text. Utilizing the KHATT dataset for comprehensive training and evaluation, implementing rigorous pre-processing steps to enhance data quality. Central to this methodology is the Res-Net152 architecture for feature extraction, which has proven highly effective. This approach achieved remarkable results, with a character error rate of approximately 2.96% and an accuracy of 97.04% on the testing dataset. These results significantly outperform the previous method, representing a substantial advancement in the field of Arabic handwriting recognition. The study demonstrates the potential of deep learning models in overcoming the unique challenges posed by Arabic script, paving the way for further improvements and applications.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":" 41","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141832985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
: Recommendation systems play a crucial role in modern information retrieval, e-commerce, and personalized content delivery. This paper provides a comprehensive review of recommendation systems, covering key concepts, methodologies, and applications. It examines different types of recommendation algorithms, including collaborative filtering, content-based filtering, and hybrid approaches, along with evaluation metrics and challenges. Our automatic recommendations and pricing system application aimed at assisting users in selecting and purchasing the optimal PC or laptop aligns with the modern demand for streamlined technology decisions. This innovative app serves as a comprehensive tool, harnessing user input to curate personalized recommendations while offering access to an extensive database of computer products. Our main contribution is improving the traditional collaborative filtering approach with a novel weighting scheme. We introduce a dynamic weighting mechanism that considers the recency and relevance of interactions to improve the accuracy and personalization of recommendations. Our recommendation systems platform, implementing a novel weighting scheme, observed a 20% increase in click-through rates (CTR) due to more relevant product recommendations. The paper also discusses emerging upcoming patterns and directions in recommendation system research.
{"title":"Automatic recommendations and pricing system for computing devices","authors":"Mohamed Refaat Mohaned Abdellah, Hossam Gamal, Asaad Hassan","doi":"10.21608/ijt.2024.290773.1051","DOIUrl":"https://doi.org/10.21608/ijt.2024.290773.1051","url":null,"abstract":": Recommendation systems play a crucial role in modern information retrieval, e-commerce, and personalized content delivery. This paper provides a comprehensive review of recommendation systems, covering key concepts, methodologies, and applications. It examines different types of recommendation algorithms, including collaborative filtering, content-based filtering, and hybrid approaches, along with evaluation metrics and challenges. Our automatic recommendations and pricing system application aimed at assisting users in selecting and purchasing the optimal PC or laptop aligns with the modern demand for streamlined technology decisions. This innovative app serves as a comprehensive tool, harnessing user input to curate personalized recommendations while offering access to an extensive database of computer products. Our main contribution is improving the traditional collaborative filtering approach with a novel weighting scheme. We introduce a dynamic weighting mechanism that considers the recency and relevance of interactions to improve the accuracy and personalization of recommendations. Our recommendation systems platform, implementing a novel weighting scheme, observed a 20% increase in click-through rates (CTR) due to more relevant product recommendations. The paper also discusses emerging upcoming patterns and directions in recommendation system research.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"65 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141682308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-01DOI: 10.21608/ijt.2024.293170.1054
M. ELBouridy, A. EL-Batouty, Marwa E Samara, Wael Abouelwafa Ahmed, Mohamed Massoud
: One of the most influential factors in preserving a person's life is the late detection of heart disease, as cardiovascular disease is considered one of the biggest risks that lead to death. Cholesterol level, age, gender, as well as blood sugar level and heart rate are considered among the most influential factors in heart disease. The accurate diagnosis of all these diseases depends on the experience and skill of the treating physician. Many researchers have intended to use automated methods to diagnose diseases without relying on the expertise of doctors. In this research, the researchers present a proposal based on deep learning (DL) using the distinctive features of some factors affecting heart disease. Therefore, magnification techniques were used to diagnose whether the patient is at risk for cardiovascular disease. Bloody or not. The research resulted in progress, as accuracy in diagnosis reached 90.088%.
{"title":"A proposed heart disease diagnosis based on Deep learning.","authors":"M. ELBouridy, A. EL-Batouty, Marwa E Samara, Wael Abouelwafa Ahmed, Mohamed Massoud","doi":"10.21608/ijt.2024.293170.1054","DOIUrl":"https://doi.org/10.21608/ijt.2024.293170.1054","url":null,"abstract":": One of the most influential factors in preserving a person's life is the late detection of heart disease, as cardiovascular disease is considered one of the biggest risks that lead to death. Cholesterol level, age, gender, as well as blood sugar level and heart rate are considered among the most influential factors in heart disease. The accurate diagnosis of all these diseases depends on the experience and skill of the treating physician. Many researchers have intended to use automated methods to diagnose diseases without relying on the expertise of doctors. In this research, the researchers present a proposal based on deep learning (DL) using the distinctive features of some factors affecting heart disease. Therefore, magnification techniques were used to diagnose whether the patient is at risk for cardiovascular disease. Bloody or not. The research resulted in progress, as accuracy in diagnosis reached 90.088%.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"30 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141840982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-01DOI: 10.21608/ijt.2024.293947.1055
Aya ElSayed, Noha A. Hikal, Nehal A. Sakr, Ali E. Takieldeen
: The present study introduces a novel hybrid deep learning model, leveraging the synergies inherent in the amalgamation of ResNet50 and DenseNet121 architectures. This fusion aims to effectively tackle the formidable task of detecting spoof attacks. Spoof attacks pose a significant threat to digital systems and networks, where adversaries attempt to deceive systems by impersonating legitimate users or sources. The proposed hybrid model aims to enhance detection accuracy and robustness against various spoof attacks by leveraging the complementary features of ResNet50 and DenseNet121. Integrating these architectures creates a unified framework that effectively captures local and global input data features, enabling more comprehensive detection capabilities. The problem of detecting spoofing attacks is stated as a classification task, and we train the hybrid model using large-scale datasets comprising fake and real data samples. The experimental results illustrate the superior performance of the proposed hybrid model in comparison to individual SVM, KNN, CNN, and RNN models, highlighting its efficacy in mitigating the risks associated with spoof attacks in digital systems and networks.
{"title":"Secure Facial Verification: A hybrid model for detecting Spoof Attacks with ResNet50-DenseNet121","authors":"Aya ElSayed, Noha A. Hikal, Nehal A. Sakr, Ali E. Takieldeen","doi":"10.21608/ijt.2024.293947.1055","DOIUrl":"https://doi.org/10.21608/ijt.2024.293947.1055","url":null,"abstract":": The present study introduces a novel hybrid deep learning model, leveraging the synergies inherent in the amalgamation of ResNet50 and DenseNet121 architectures. This fusion aims to effectively tackle the formidable task of detecting spoof attacks. Spoof attacks pose a significant threat to digital systems and networks, where adversaries attempt to deceive systems by impersonating legitimate users or sources. The proposed hybrid model aims to enhance detection accuracy and robustness against various spoof attacks by leveraging the complementary features of ResNet50 and DenseNet121. Integrating these architectures creates a unified framework that effectively captures local and global input data features, enabling more comprehensive detection capabilities. The problem of detecting spoofing attacks is stated as a classification task, and we train the hybrid model using large-scale datasets comprising fake and real data samples. The experimental results illustrate the superior performance of the proposed hybrid model in comparison to individual SVM, KNN, CNN, and RNN models, highlighting its efficacy in mitigating the risks associated with spoof attacks in digital systems and networks.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"21 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141848853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-20DOI: 10.21608/ijt.2024.269831.1043
Mohamed Bakr, Noha Korany
: The technological scene is developing towards employing tiny-sized devices. For a variety of functions that encompass sensing, identification, and decision-making. These devices are restricted in their resources to be secure. The National Institute of Standards and Technology (NIST) has published the Ascon, a standard lightweight cryptography (LWC) algorithm for data gathered by restricted devices in 2023. The fundamental function of Ascon is the permutation function. The main core of the permutation function is the substitution box (S-box). This paper proposes an S-box based on chaotic systems using coupled map lattices (CML), which agrees with LWC algorithms due to its low complexity. The proposed S-box suggests high performance and security for restricted devices. The security robustness is tested using various cryptographic criteria, such as nonlinearity, strict avalanche criteria, and differential approximation probabilities. The proposed S-box approaches significant resistance to both linear and differential cryptanalysis. So, it could be replacing the substitution and linear diffusion layers of Ascon permutation to improve the nonlinearity and randomness.
{"title":"Nonlinearity Improvement for Ascon Substitution Box","authors":"Mohamed Bakr, Noha Korany","doi":"10.21608/ijt.2024.269831.1043","DOIUrl":"https://doi.org/10.21608/ijt.2024.269831.1043","url":null,"abstract":": The technological scene is developing towards employing tiny-sized devices. For a variety of functions that encompass sensing, identification, and decision-making. These devices are restricted in their resources to be secure. The National Institute of Standards and Technology (NIST) has published the Ascon, a standard lightweight cryptography (LWC) algorithm for data gathered by restricted devices in 2023. The fundamental function of Ascon is the permutation function. The main core of the permutation function is the substitution box (S-box). This paper proposes an S-box based on chaotic systems using coupled map lattices (CML), which agrees with LWC algorithms due to its low complexity. The proposed S-box suggests high performance and security for restricted devices. The security robustness is tested using various cryptographic criteria, such as nonlinearity, strict avalanche criteria, and differential approximation probabilities. The proposed S-box approaches significant resistance to both linear and differential cryptanalysis. So, it could be replacing the substitution and linear diffusion layers of Ascon permutation to improve the nonlinearity and randomness.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":" 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140388928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-24DOI: 10.21608/ijt.2024.342832
Mohamed Badawi, Al Nagar ,E Al Nagar, Mansour, R Mansour, R, Ibrahim ,Kh Ibrahim ,Kh, Nada Hegazy, Safa Elaskary
: Nowadays, Smart Glass emerges as a potential aid for individuals with visual impairments, offering the promise of enhanced quality of life. Designed for those seeking independent navigation with a sense of social ease and security, the concept revolves around the idea that visually impaired individuals prefer inconspicuous assistance tools. This paper delves into the significant advancements within wearable electronics, spot-lighting additional features. This innovative glass offers a multifaceted solution for individuals with visual impairments, providing assistance in diverse scenarios. Beyond aiding in the reading of scripts, they excel at distinguishing between currencies, enabling users to navigate financial transactions with ease. The glasses also enhance color recognition, allowing wearers to perceive and appreciate the vibrant spectrum of the world around them. Additionally, the incorporation of obstacle detection technology ensures a heightened sense of safety by alerting users when they are in proximity to potential hazards. Furthermore, the glasses feature advanced facial recognition capabilities, contributing to a more inclusive and socially connected experience by detecting faces and fostering seamless interactions.
{"title":"Smart Bionic Vision: An Assistive Device System for the Vis-ually Impaired Using Artificial Intelligence","authors":"Mohamed Badawi, Al Nagar ,E Al Nagar, Mansour, R Mansour, R, Ibrahim ,Kh Ibrahim ,Kh, Nada Hegazy, Safa Elaskary","doi":"10.21608/ijt.2024.342832","DOIUrl":"https://doi.org/10.21608/ijt.2024.342832","url":null,"abstract":": Nowadays, Smart Glass emerges as a potential aid for individuals with visual impairments, offering the promise of enhanced quality of life. Designed for those seeking independent navigation with a sense of social ease and security, the concept revolves around the idea that visually impaired individuals prefer inconspicuous assistance tools. This paper delves into the significant advancements within wearable electronics, spot-lighting additional features. This innovative glass offers a multifaceted solution for individuals with visual impairments, providing assistance in diverse scenarios. Beyond aiding in the reading of scripts, they excel at distinguishing between currencies, enabling users to navigate financial transactions with ease. The glasses also enhance color recognition, allowing wearers to perceive and appreciate the vibrant spectrum of the world around them. Additionally, the incorporation of obstacle detection technology ensures a heightened sense of safety by alerting users when they are in proximity to potential hazards. Furthermore, the glasses feature advanced facial recognition capabilities, contributing to a more inclusive and socially connected experience by detecting faces and fostering seamless interactions.","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"75 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140434278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-08DOI: 10.21608/ijt.2024.340441
sherif mahgoub, Mohamed Ashour, Mohamed Yakout, Eman AbdElhalim
{"title":"Traffic Classification in Software Defined Networks based on Machine Learning Algorithms","authors":"sherif mahgoub, Mohamed Ashour, Mohamed Yakout, Eman AbdElhalim","doi":"10.21608/ijt.2024.340441","DOIUrl":"https://doi.org/10.21608/ijt.2024.340441","url":null,"abstract":"","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"12 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.21608/ijt.2024.262285.1042
Mohamed Massoud, Gehan Mahmoud, Waheed Ali, Wael Ahmed
Hand spasticity poses a significant challenge for stroke survivors, impacting hand functionality and hindering daily activities . The study introduces a smart rehabilitation system engineered for post-stroke hand spasticity. Comprising four units includes biometric measurement gloves, rehabilitation gloves, a camera, a telecom unit, and a computer unit . Biometric measurement gloves with sensors measure patient features. Data inputs include biometric measurements and cam-era-captured images. Computer programs consist of a clinical biometric program and a CNN program, specifically ResNet50 architecture . The telecom unit facilitates communication between the computer unit and rehabilittion gloves, doctor section, and patient section. The smart rehabilitation system offers advantages such as user-friendly operation, cost-effectiveness, elimination of physical visits to rehabilitation centers, and exceptional accuracy with a 99% validation accuracy rate and 0.0053 validation loss in the CNN framework. The clinical biometric program is used to analyze programs with high accuracy . This study presents an innovative rehabilitation system. It includes biometric measurement gloves for patient assessment and rehabilitation gloves for hand exercises. Two programs, a clinical biometric program, and an intelligent CNN-based program, diagnose and therapies based on biometric data and image analysis. The mobile application communicates be-tween the system
{"title":"Revolutionizing Stroke Rehabilitation: Dynamic Glove-Based Rehabilitation System Empowered by CNN for Spastic Hands.","authors":"Mohamed Massoud, Gehan Mahmoud, Waheed Ali, Wael Ahmed","doi":"10.21608/ijt.2024.262285.1042","DOIUrl":"https://doi.org/10.21608/ijt.2024.262285.1042","url":null,"abstract":"Hand spasticity poses a significant challenge for stroke survivors, impacting hand functionality and hindering daily activities . The study introduces a smart rehabilitation system engineered for post-stroke hand spasticity. Comprising four units includes biometric measurement gloves, rehabilitation gloves, a camera, a telecom unit, and a computer unit . Biometric measurement gloves with sensors measure patient features. Data inputs include biometric measurements and cam-era-captured images. Computer programs consist of a clinical biometric program and a CNN program, specifically ResNet50 architecture . The telecom unit facilitates communication between the computer unit and rehabilittion gloves, doctor section, and patient section. The smart rehabilitation system offers advantages such as user-friendly operation, cost-effectiveness, elimination of physical visits to rehabilitation centers, and exceptional accuracy with a 99% validation accuracy rate and 0.0053 validation loss in the CNN framework. The clinical biometric program is used to analyze programs with high accuracy . This study presents an innovative rehabilitation system. It includes biometric measurement gloves for patient assessment and rehabilitation gloves for hand exercises. Two programs, a clinical biometric program, and an intelligent CNN-based program, diagnose and therapies based on biometric data and image analysis. The mobile application communicates be-tween the system","PeriodicalId":517010,"journal":{"name":"International Journal of Telecommunications","volume":"164 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140462734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：