In this article, through the utilization of the extended hyperbolic function method (EHFM), we derive the exact solutions for the modified Benjamin–Bona–Mahoni (mBBM) and Eckhaus equations. The EHFM introduces various novel solutions, including dark, bright, singular, singular-periodic, kink solitons, and some rational function solutions. Furthermore, due to a deeper understanding of the physical structure of these solutions, we used Maple and MATLAB software to generate 3D and contour representations of some solutions. The obtained outcomes are innovative regarding the considered equations, and the findings demonstrate that the approach is practical, concise, and direct, which can be applied to describe other physical phenomena.
{"title":"Dynamical exploration of novel soliton solutions of the modified Benjamin–Bona–Mahoni and Eckhaus equations based on the extended hyperbolic function method","authors":"Mostafa Eslami , Yasin Asghari , Mashallah Matinfar , Hadi Rezazadeh","doi":"10.1016/j.aej.2025.02.106","DOIUrl":"10.1016/j.aej.2025.02.106","url":null,"abstract":"<div><div>In this article, through the utilization of the extended hyperbolic function method (EHFM), we derive the exact solutions for the modified Benjamin–Bona–Mahoni (mBBM) and Eckhaus equations. The EHFM introduces various novel solutions, including dark, bright, singular, singular-periodic, kink solitons, and some rational function solutions. Furthermore, due to a deeper understanding of the physical structure of these solutions, we used Maple and MATLAB software to generate 3D and contour representations of some solutions. The obtained outcomes are innovative regarding the considered equations, and the findings demonstrate that the approach is practical, concise, and direct, which can be applied to describe other physical phenomena.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"123 ","pages":"Pages 46-54"},"PeriodicalIF":6.2,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-22DOI: 10.1016/j.aej.2025.03.030
Tunahan Gokcimen , Bihter Das
To address the escalating demand for secure and trustworthy interactions with Large Language Models (LLMs), this study introduces a pioneering security framework that mitigates critical vulnerabilities, including injection attacks, hallucinations, and data privacy breaches. By incorporating advanced technologies such as VectorDB, Kernel, and Retrieval-Augmented Generation (RAG) within a cross-LLM architecture, the system delivers enhanced resilience and adaptability to adversarial scenarios. Comprehensive evaluations across leading models—including PaLM, Llama, GPT-3.5, GPT-4, Gemini, and GPT-4o—reveal the system’s exceptional performance, achieving a 98 % accuracy in eligibility scoring and outperforming conventional models in both reliability and security. This study underscores the significance of a multi-layered defense mechanism that not only detects and neutralizes threats but also ensures ethical, accurate, and contextually relevant responses. The novel cross-LLM strategy enhances system robustness by leveraging the strengths of multiple models, minimizing inconsistencies and reinforcing output integrity. With its adaptability to emerging linguistic manipulation techniques and compliance with strict ethical standards, the proposed framework establishes a secure, scalable ecosystem for LLM applications. The findings promise transformative impacts across domains such as cybersecurity, multilingual processing, and adaptive threat detection, paving the way for safer and more reliable language model deployments.
{"title":"A novel system for strengthening security in large language models against hallucination and injection attacks with effective strategies","authors":"Tunahan Gokcimen , Bihter Das","doi":"10.1016/j.aej.2025.03.030","DOIUrl":"10.1016/j.aej.2025.03.030","url":null,"abstract":"<div><div>To address the escalating demand for secure and trustworthy interactions with Large Language Models (LLMs), this study introduces a pioneering security framework that mitigates critical vulnerabilities, including injection attacks, hallucinations, and data privacy breaches. By incorporating advanced technologies such as VectorDB, Kernel, and Retrieval-Augmented Generation (RAG) within a cross-LLM architecture, the system delivers enhanced resilience and adaptability to adversarial scenarios. Comprehensive evaluations across leading models—including PaLM, Llama, GPT-3.5, GPT-4, Gemini, and GPT-4o—reveal the system’s exceptional performance, achieving a 98 % accuracy in eligibility scoring and outperforming conventional models in both reliability and security. This study underscores the significance of a multi-layered defense mechanism that not only detects and neutralizes threats but also ensures ethical, accurate, and contextually relevant responses. The novel cross-LLM strategy enhances system robustness by leveraging the strengths of multiple models, minimizing inconsistencies and reinforcing output integrity. With its adaptability to emerging linguistic manipulation techniques and compliance with strict ethical standards, the proposed framework establishes a secure, scalable ecosystem for LLM applications. The findings promise transformative impacts across domains such as cybersecurity, multilingual processing, and adaptive threat detection, paving the way for safer and more reliable language model deployments.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"123 ","pages":"Pages 71-90"},"PeriodicalIF":6.2,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-22DOI: 10.1016/j.aej.2025.03.044
Kai Wang , Wensheng Jia
In this paper, the chaotic whale slime mould algorithm (CWSMA) is designed to solve the generalized Nash equilibrium problem (GNEP). First, the GNEP is converted into a non-linear equation system problem using the Karush–Kuhn–Tucker (KKT) condition and the “min” function. Compared to the classical approach, the transformation process does not require the functions to be quadratically differentiable and strongly convex. The CWSMA is proposed by introducing tent mapping, levy flight strategy and the idea of whale optimization algorithm into the slime mould algorithm (SMA), which has the advantages of higher population diversity, faster convergence, less chance of falling into local optimums, and does not depend on the choice of initial points. Furthermore, the convergence analysis of the CWSMA is given by using Markov processes. Finally, several numerical simulation experiments show that the CWSMA is superior to the SMA, the coevolutionary immune quantum particle swarm optimization algorithm, the projection algorithm and the descent algorithm under certain conditions. The CWSMA not only solves the GNEP effectively, but also has better population diversity, global convergence, and off-line performance.
{"title":"A new intelligent algorithm for solving generalized Nash equilibrium problem","authors":"Kai Wang , Wensheng Jia","doi":"10.1016/j.aej.2025.03.044","DOIUrl":"10.1016/j.aej.2025.03.044","url":null,"abstract":"<div><div>In this paper, the chaotic whale slime mould algorithm (CWSMA) is designed to solve the generalized Nash equilibrium problem (GNEP). First, the GNEP is converted into a non-linear equation system problem using the Karush–Kuhn–Tucker (KKT) condition and the “min” function. Compared to the classical approach, the transformation process does not require the functions to be quadratically differentiable and strongly convex. The CWSMA is proposed by introducing tent mapping, levy flight strategy and the idea of whale optimization algorithm into the slime mould algorithm (SMA), which has the advantages of higher population diversity, faster convergence, less chance of falling into local optimums, and does not depend on the choice of initial points. Furthermore, the convergence analysis of the CWSMA is given by using Markov processes. Finally, several numerical simulation experiments show that the CWSMA is superior to the SMA, the coevolutionary immune quantum particle swarm optimization algorithm, the projection algorithm and the descent algorithm under certain conditions. The CWSMA not only solves the GNEP effectively, but also has better population diversity, global convergence, and off-line performance.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"123 ","pages":"Pages 17-28"},"PeriodicalIF":6.2,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-22DOI: 10.1016/j.aej.2025.03.035
Abdul-Majid Wazwaz , Samir A. El-Tantawy , L.S. El-Sherif , Amnah S. Al-Johani , Haifa A. Alyousef
Several scientific fields, including mechanical fluids, plasmas, and solids, use higher-dimensional Boussinesq-type equations to model various nonlinear phenomena, such as solitary and shock waves, as well as lump waves. Motivated by these applications, this investigation focuses on studying and analyzing four novel integrable higher-dimensional Boussinesq and Boussinesq-type equations, drawing from the many and varied applications of this family of evolution equations. To the best of the author’s knowledge, these four models are constructed and introduced for the first time. We first check the complete integrability for the four suggested models via Painlevé analysis. Next, we employ the simplified Hirota’s method (SHM) to solve the four proposed models and derive multiple soliton solutions. Our results show that the simplified Hirota’s approach is efficient and robust for solving these equations. In addition, we use symbolic computation with Maple to explicitly derive two distinct categories of lump solutions for each equation. We numerically investigate all derived solitary and lump solutions to understand the dynamical behavior of these waves. Note that this approach can also derive other lump solutions. This study’s findings should benefit many researchers in fluid and plasma physics and analytical engineering should benefit from the findings of this study.
{"title":"Multiple soliton and lump solutions to a variety of novel integrable multi-dimensional Boussinesq-type equations","authors":"Abdul-Majid Wazwaz , Samir A. El-Tantawy , L.S. El-Sherif , Amnah S. Al-Johani , Haifa A. Alyousef","doi":"10.1016/j.aej.2025.03.035","DOIUrl":"10.1016/j.aej.2025.03.035","url":null,"abstract":"<div><div>Several scientific fields, including mechanical fluids, plasmas, and solids, use higher-dimensional Boussinesq-type equations to model various nonlinear phenomena, such as solitary and shock waves, as well as lump waves. Motivated by these applications, this investigation focuses on studying and analyzing four novel integrable higher-dimensional Boussinesq and Boussinesq-type equations, drawing from the many and varied applications of this family of evolution equations. To the best of the author’s knowledge, these four models are constructed and introduced for the first time. We first check the complete integrability for the four suggested models via Painlevé analysis. Next, we employ the simplified Hirota’s method (SHM) to solve the four proposed models and derive multiple soliton solutions. Our results show that the simplified Hirota’s approach is efficient and robust for solving these equations. In addition, we use symbolic computation with Maple to explicitly derive two distinct categories of lump solutions for each equation. We numerically investigate all derived solitary and lump solutions to understand the dynamical behavior of these waves. Note that this approach can also derive other lump solutions. This study’s findings should benefit many researchers in fluid and plasma physics and analytical engineering should benefit from the findings of this study.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"123 ","pages":"Pages 1-16"},"PeriodicalIF":6.2,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143685563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The modeling and prediction of oil prices are very important tasks. However, the predictions of oil price by traditional models are not very effective. The challenge comes from the nonlinear and non-stationary dynamics of oil prices, and they are also heavily correlated with global economic condition and financial fluctuations. This study employs a dual-stage attention-based recurrent neural network (DA-RNN) for oil price forecasting. The DA-RNN architecture includes both an encoder and a decoder. The encoder features an input attention mechanism designed to adaptively identify and choose significant and pertinent driving series. The decoder incorporates a temporal attention mechanism to obtain long-term dependencies of the encoded inputs. The dual-stage attention mechanism of DA-RNN enables both input selection and temporal focus, allowing the model to adaptively choose important and relevant driving series while capturing long-term temporal dependencies. Empirical results indicate that DA-RNN achieved lowest prediction errors, for example, RMSE values of approximately 2.2 for WTI, 2.4 for Dubai, and 2.3 for Brent crude oil prices, which reduces about 30 % error compared to other models. These findings clearly demonstrated that the DA-RNN model outperforms traditional econometric methods and machine learning models, highlighting its potential as a powerful tool for energy market predictions.
{"title":"Intelligent dual-stage attention-based deep networks for energy market predictions","authors":"Shian-Chang Huang , Cheng-Feng Wu , Kuan-Chieh Chen , Meng-Chen Lin , Chei-Chang Chiou","doi":"10.1016/j.aej.2025.03.031","DOIUrl":"10.1016/j.aej.2025.03.031","url":null,"abstract":"<div><div>The modeling and prediction of oil prices are very important tasks. However, the predictions of oil price by traditional models are not very effective. The challenge comes from the nonlinear and non-stationary dynamics of oil prices, and they are also heavily correlated with global economic condition and financial fluctuations. This study employs a dual-stage attention-based recurrent neural network (DA-RNN) for oil price forecasting. The DA-RNN architecture includes both an encoder and a decoder. The encoder features an input attention mechanism designed to adaptively identify and choose significant and pertinent driving series. The decoder incorporates a temporal attention mechanism to obtain long-term dependencies of the encoded inputs. The dual-stage attention mechanism of DA-RNN enables both input selection and temporal focus, allowing the model to adaptively choose important and relevant driving series while capturing long-term temporal dependencies. Empirical results indicate that DA-RNN achieved lowest prediction errors, for example, RMSE values of approximately 2.2 for WTI, 2.4 for Dubai, and 2.3 for Brent crude oil prices, which reduces about 30 % error compared to other models. These findings clearly demonstrated that the DA-RNN model outperforms traditional econometric methods and machine learning models, highlighting its potential as a powerful tool for energy market predictions.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 625-644"},"PeriodicalIF":6.2,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-21DOI: 10.1016/j.aej.2025.03.059
Mohan Liu , Qingqing Liu , Fengxia Liu
This study presents a novel electrochemical sensor for theophylline quantification based on a graphene-ZnO nanocomposite. To create the sensor, the glassy carbon electrode was altered by incorporating a carefully balanced mixture of 30 % graphene and 70 % ZnO. Characterization techniques, including SEM, XRD, and FTIR, confirmed the successful synthesis and integration of the nanocomposite. The sensor showed outstanding analytical capabilities, with two distinct linear ranges from 0.1 to to 100 μM, and a detection limit of 0.03 μM. Through cyclic voltammetry experiments, it was determined that the oxidation process was diffusion-controlled, and the best response was achieved at pH 7.0. The sensor showed strong selectivity towards typical interferents, with the ability to withstand concentrations of up to 500 μM for certain substances. Practical applicability was validated through analysis of pharmaceutical formulations and spiked biological samples, yielding recoveries of 98.5–101.2 % and 96.8–103.1 %, respectively. The sensor showed good reproducibility (RSD < 4 %) and retained 95.6 %, indicating promising potential for routine theophylline monitoring in clinical and pharmaceutical settings.
{"title":"Quantification of theophylline in anti-asthmatic drugs via graphene-ZnO based sensors","authors":"Mohan Liu , Qingqing Liu , Fengxia Liu","doi":"10.1016/j.aej.2025.03.059","DOIUrl":"10.1016/j.aej.2025.03.059","url":null,"abstract":"<div><div>This study presents a novel electrochemical sensor for theophylline quantification based on a graphene-ZnO nanocomposite. To create the sensor, the glassy carbon electrode was altered by incorporating a carefully balanced mixture of 30 % graphene and 70 % ZnO. Characterization techniques, including SEM, XRD, and FTIR, confirmed the successful synthesis and integration of the nanocomposite. The sensor showed outstanding analytical capabilities, with two distinct linear ranges from 0.1 to to 100 μM, and a detection limit of 0.03 μM. Through cyclic voltammetry experiments, it was determined that the oxidation process was diffusion-controlled, and the best response was achieved at pH 7.0. The sensor showed strong selectivity towards typical interferents, with the ability to withstand concentrations of up to 500 μM for certain substances. Practical applicability was validated through analysis of pharmaceutical formulations and spiked biological samples, yielding recoveries of 98.5–101.2 % and 96.8–103.1 %, respectively. The sensor showed good reproducibility (RSD < 4 %) and retained 95.6 %, indicating promising potential for routine theophylline monitoring in clinical and pharmaceutical settings.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 645-654"},"PeriodicalIF":6.2,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-21DOI: 10.1016/j.aej.2025.03.056
Shuai Wang
This study was conducted on developing a colorimetric aptasensor based on bimetallic Zn-Fe metal-organic framework (MOF) nanostructures and an application for the determination of nandrolone (ND) as doping anabolic steroid agents in sports. The aptasensor was synthesized by functionalizing Zn-Fe MOF nanoparticles with amino groups, followed by the conjugation of carboxyl-activated aptamers specific to ND. The performance of the aptasensor was assessed through a TMB-H2O2 colorimetric assay, highlighting great selectivity and sensitivity for ND’s detection range of 0.005–1000 µM and a limit of detection of 0.85 nM. The comparison of the performance of the current study and the other ND sensor reports exhibited the supreme sensitivity and dynamic range of the Zn-Fe MOF based aptasensor. The aptasensor showed appropriate stability, retaining 96.77 % of its initial response after 21 days and 93.13 % after 30 days, indicating its appropriate long-term applications. In the examination of sensor capability for the determination of ND in human serum and urine samples, results displayed excellent recovery rates from 97.00 % to 99.30 % and low RSD values (less than 4.42 %), highlighting its reliability for doping control and clinical analyses. The incorporation of functionalized bimetallic MOF nanostructures can promote the biosensing capabilities, providing a robust and cost-effective solution for monitoring ND levels in biological and medicinal samples.
{"title":"Innovative colorimetric sensor for detection of nandrolone as a doping agent in sports using MOF nanostructures","authors":"Shuai Wang","doi":"10.1016/j.aej.2025.03.056","DOIUrl":"10.1016/j.aej.2025.03.056","url":null,"abstract":"<div><div>This study was conducted on developing a colorimetric aptasensor based on bimetallic Zn-Fe metal-organic framework (MOF) nanostructures and an application for the determination of nandrolone (ND) as doping anabolic steroid agents in sports. The aptasensor was synthesized by functionalizing Zn-Fe MOF nanoparticles with amino groups, followed by the conjugation of carboxyl-activated aptamers specific to ND. The performance of the aptasensor was assessed through a TMB-H<sub>2</sub>O<sub>2</sub> colorimetric assay, highlighting great selectivity and sensitivity for ND’s detection range of 0.005–1000 µM and a limit of detection of 0.85 nM. The comparison of the performance of the current study and the other ND sensor reports exhibited the supreme sensitivity and dynamic range of the Zn-Fe MOF based aptasensor. The aptasensor showed appropriate stability, retaining 96.77 % of its initial response after 21 days and 93.13 % after 30 days, indicating its appropriate long-term applications. In the examination of sensor capability for the determination of ND in human serum and urine samples, results displayed excellent recovery rates from 97.00 % to 99.30 % and low RSD values (less than 4.42 %), highlighting its reliability for doping control and clinical analyses. The incorporation of functionalized bimetallic MOF nanostructures can promote the biosensing capabilities, providing a robust and cost-effective solution for monitoring ND levels in biological and medicinal samples.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 605-614"},"PeriodicalIF":6.2,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-21DOI: 10.1016/j.aej.2025.03.005
Jiongen Xiao , Zifeng Zhang
Visual Question Answering (VQA) shows great potential in educational fields like textbook analysis, classroom interaction, and gamified learning. However, existing VQA systems face significant challenges in addressing the unique complexities of educational scenarios. On one hand, most models lack the ability to dynamically comprehend multimodal contexts, making it difficult to meet the diverse semantic demands of educational tasks. On the other hand, many methods fail to fully leverage the reasoning capabilities of large language models (LLMs), resulting in limited performance on knowledge-driven tasks. To overcome these challenges, we propose a novel VQA framework, EduVQA, specifically designed for educational scenarios. EduVQA incorporates a dynamic context selection mechanism and a pre-answer generation module to effectively manage the complexity of multimodal data in educational contexts. Furthermore, by integrating a fine-tuned large language model, EduVQA significantly enhances the understanding and reasoning needed for complex educational questions. Specifically, EduVQA dynamically filters context information relevant to the questions to reduce noise and employs a multi-level pre-answer generation module, combined with external knowledge bases, to provide precise guidance for answer generation. Experimental results show that EduVQA significantly outperforms state-of-the-art models on the OK-VQA and A-OKVQA datasets, excelling in tasks requiring knowledge reasoning and logical analysis.
{"title":"EduVQA: A multimodal Visual Question Answering framework for smart education","authors":"Jiongen Xiao , Zifeng Zhang","doi":"10.1016/j.aej.2025.03.005","DOIUrl":"10.1016/j.aej.2025.03.005","url":null,"abstract":"<div><div>Visual Question Answering (VQA) shows great potential in educational fields like textbook analysis, classroom interaction, and gamified learning. However, existing VQA systems face significant challenges in addressing the unique complexities of educational scenarios. On one hand, most models lack the ability to dynamically comprehend multimodal contexts, making it difficult to meet the diverse semantic demands of educational tasks. On the other hand, many methods fail to fully leverage the reasoning capabilities of large language models (LLMs), resulting in limited performance on knowledge-driven tasks. To overcome these challenges, we propose a novel VQA framework, EduVQA, specifically designed for educational scenarios. EduVQA incorporates a dynamic context selection mechanism and a pre-answer generation module to effectively manage the complexity of multimodal data in educational contexts. Furthermore, by integrating a fine-tuned large language model, EduVQA significantly enhances the understanding and reasoning needed for complex educational questions. Specifically, EduVQA dynamically filters context information relevant to the questions to reduce noise and employs a multi-level pre-answer generation module, combined with external knowledge bases, to provide precise guidance for answer generation. Experimental results show that EduVQA significantly outperforms state-of-the-art models on the OK-VQA and A-OKVQA datasets, excelling in tasks requiring knowledge reasoning and logical analysis.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 615-624"},"PeriodicalIF":6.2,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-20DOI: 10.1016/j.aej.2025.03.026
Deep Singh , Harpreet Kaur , Chaman Verma , Neerendra Kumar , Zoltán Illés
The security of 3-D images is an important research problem that has to be resolved. Due to the more complex structure of 3-D images, the encryption of these images is quite different from 1-D and 2-D images. A three-stage novel image encryption technique based on chaotic maps is presented in this paper in which a 3-D image is firstly converted to a similar image format as that of 2-D images before encryption. The initial conditions of the chaotic system are generated by using the SHA-256 function on the coordinate matrix of the plaintext. Initially, a logistic map is utilized to scramble and add random points to the coordinate values of vertices of a 3-D image. The coordinate values are then confused and diffused in the second stage by using three sequences generated through the logistic-dynamic coupled logistic map lattice (LDCML) model. This stage also involves splitting of floating point data into integer and fractional parts. The integer part is diffused whereas the fractional part is scrambled during this process. In the third stage, the confusion is performed by using a tent map among the coordinate points. This process enhances the robustness, integrity, and confidentiality of 3-D images and ensures protection against unauthorized access. The proposed encryption procedure achieves values of correlation close to zero along x,y,z- directions, NPCR of 100%, UACI of 33.37%, and entropy value of 7.9993 which demonstrates its robust security. The time analysis shows that our technique improves efficiency and lowers computational costs by processing data in a lesser time. The security and statistical analysis concludes that the proposed encryption algorithm can withstand various conventional attacks.
{"title":"A novel 3-D image encryption algorithm based on SHA-256 and chaos theory","authors":"Deep Singh , Harpreet Kaur , Chaman Verma , Neerendra Kumar , Zoltán Illés","doi":"10.1016/j.aej.2025.03.026","DOIUrl":"10.1016/j.aej.2025.03.026","url":null,"abstract":"<div><div>The security of 3-D images is an important research problem that has to be resolved. Due to the more complex structure of 3-D images, the encryption of these images is quite different from 1-D and 2-D images. A three-stage novel image encryption technique based on chaotic maps is presented in this paper in which a 3-D image is firstly converted to a similar image format as that of 2-D images before encryption. The initial conditions of the chaotic system are generated by using the SHA-256 function on the coordinate matrix of the plaintext. Initially, a logistic map is utilized to scramble and add random points to the coordinate values of vertices of a 3-D image. The coordinate values are then confused and diffused in the second stage by using three sequences generated through the logistic-dynamic coupled logistic map lattice (LDCML) model. This stage also involves splitting of floating point data into integer and fractional parts. The integer part is diffused whereas the fractional part is scrambled during this process. In the third stage, the confusion is performed by using a tent map among the coordinate points. This process enhances the robustness, integrity, and confidentiality of 3-D images and ensures protection against unauthorized access. The proposed encryption procedure achieves values of correlation close to zero along x,y,z- directions, NPCR of 100%, UACI of 33.37%, and entropy value of 7.9993 which demonstrates its robust security. The time analysis shows that our technique improves efficiency and lowers computational costs by processing data in a lesser time. The security and statistical analysis concludes that the proposed encryption algorithm can withstand various conventional attacks.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 564-577"},"PeriodicalIF":6.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-20DOI: 10.1016/j.aej.2025.03.032
Meng Wang , Zi Yang , Ruifeng Zhao , Yaoting Jiang
The integration of Internet of Things (IoT) technology in pulmonary nodule detection significantly enhances the intelligence and real-time capabilities of the detection system. Currently, lung nodule detection primarily focuses on the identification of solid nodules, but different types of lung nodules correspond to various forms of lung cancer. Multi-type detection contributes to improving the overall lung cancer detection rate and enhancing the cure rate. To achieve high sensitivity in nodule detection, targeted improvements were made to the YOLOv8 model. Firstly, the C2f_RepViTCAMF module was introduced to augment the C2f module in the backbone, thereby enhancing detection accuracy for small lung nodules and achieving a lightweight model design. Secondly, the MSCAF module was incorporated to reconstruct the feature fusion section of the model, improving detection accuracy for lung nodules of varying scales. Furthermore, the KAN network was integrated into the model. By leveraging the KAN network’s powerful nonlinear feature learning capability, detection accuracy for small lung nodules was further improved, and the model’s generalization ability was enhanced. Tests conducted on the LUNA16 dataset demonstrate that the improved model outperforms the original model as well as other mainstream models such as YOLOv9 and RT-DETR across various evaluation metrics.
{"title":"CPLOYO: A pulmonary nodule detection model with multi-scale feature fusion and nonlinear feature learning","authors":"Meng Wang , Zi Yang , Ruifeng Zhao , Yaoting Jiang","doi":"10.1016/j.aej.2025.03.032","DOIUrl":"10.1016/j.aej.2025.03.032","url":null,"abstract":"<div><div>The integration of Internet of Things (IoT) technology in pulmonary nodule detection significantly enhances the intelligence and real-time capabilities of the detection system. Currently, lung nodule detection primarily focuses on the identification of solid nodules, but different types of lung nodules correspond to various forms of lung cancer. Multi-type detection contributes to improving the overall lung cancer detection rate and enhancing the cure rate. To achieve high sensitivity in nodule detection, targeted improvements were made to the YOLOv8 model. Firstly, the C2f_RepViTCAMF module was introduced to augment the C2f module in the backbone, thereby enhancing detection accuracy for small lung nodules and achieving a lightweight model design. Secondly, the MSCAF module was incorporated to reconstruct the feature fusion section of the model, improving detection accuracy for lung nodules of varying scales. Furthermore, the KAN network was integrated into the model. By leveraging the KAN network’s powerful nonlinear feature learning capability, detection accuracy for small lung nodules was further improved, and the model’s generalization ability was enhanced. Tests conducted on the LUNA16 dataset demonstrate that the improved model outperforms the original model as well as other mainstream models such as YOLOv9 and RT-DETR across various evaluation metrics.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"122 ","pages":"Pages 578-587"},"PeriodicalIF":6.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号