检索结果-内蒙古大学图书馆

Privacy-preserving explainable AI: a survey

science China(information sciences) 2025年第1期68卷 23-56页

作者： Thanh Tam NGUYEN Thanh Trung HUYNH Zhao REN Thanh Toan NGUYEN Phi Le NGUYEN Hongzhi YIN Quoc Viet Hung NGUYEN School of Information and Communication Technology Griffith University School of Computer and Communication Sciences Ecole Polytechnique Federale de Lausanne Faculty of Mathematics and Computer Science University of Bremen Faculty of Information Technology HUTECH University Department of Computer Science Hanoi University of Science and Technology School of Electrical Engineering and Computer Science The University of Queensland

As the adoption of explainable AI(XAI) continues to expand, the urgency to address its privacy implications intensifies. Despite a growing corpus of research in AI privacy and explainability, there is little attention on privacy-preserving model explanations. This article presents the first thorough survey about privacy attacks on model explanations and their countermeasures. Our contribution to this field comprises a thorough analysis of research papers with a connected taxonomy that facilitates the categorization of privacy attacks and countermeasures based on the targeted explanations. This work also includes an initial investigation into the causes of privacy leaks. Finally, we discuss unresolved issues and prospective research directions uncovered in our analysis. This survey aims to be a valuable resource for the research community and offers clear insights for those new to this domain. To support ongoing research, we have established an online resource repository, which will be continuously updated with new and relevant findings.

关键词： privacy-preserving explainable AI privacy attacks privacy defences PrivEx PPXAI

来源：评论

学校读者我要写书评

暂无评论

Efficient and lightweight CNN model for COVID-19 diagnosis from CT and X-ray images using customized pruning and quantization techniques

引用

Neural Computing and Applications 2025年 1-20页

作者： Sriwiboon, Nattavut Department of Computer Science and Information Technology Faculty of Science and Health Technology Kalasin University Kalasin Thailand

The coronavirus disease 2019 (COVID-19) has posed significant challenges globally, with image classification becoming a critical tool for detecting COVID-19 from chest X-ray and CT images. Convolutional neural network (CNN) architectures are widely employed for this purpose, yet many previous models have struggled to balance high accuracy with low storage requirements, limiting their applicability in resource-constrained environments. This paper introduces an efficient and lightweight CNN model optimized for COVID-19 diagnosis through pruning and quantization techniques, addressing these limitations. Our approach significantly reduces storage size and parameter count while maintaining high accuracy. We customized several CNNs, particularly EfficientNetV2 and MobileNetV2, to achieve these goals. We specifically selected EfficientNetV2 and MobileNetV2 due to their superior balance of accuracy and efficiency, making them well-suited for lightweight medical applications. EfficientNetV2’s storage size has been reduced from 568 to 142 KB through customized pruning, achieving 98.9% accuracy for 2-class and 98.28% for 3-class X-ray classification, as well as 98.3% and 98.05% for 2-class and 3-class CT classification, respectively. MobileNetV2, after hybrid pruning and quantization, achieved a remarkably low storage size of 70 KB, making it highly effective for deployment in environments with limited resources. Despite a slight accuracy decrease of 0.9 to 1.58%, both models have maintained robust performance suitable for COVID-19 diagnosis. The results demonstrate that our pruning and quantization techniques are effective in developing lightweight, high-performance CNN models, making them suitable for broader applications in medical imaging where both efficiency and accuracy are critical. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Enhanced model for abstractive Arabic text summarization using natural language generation and named entity recognition

引用

Neural Computing and Applications 2025年第10期37卷 7279-7301页

作者： Essa, Nada El-Gayar, M.M. El-Daydamony, Eman M. Department of Information Technology Faculty of Computer and Information Science Mansoura University Mansoura35516 Egypt

With the rise of Arabic digital content, effective summarization methods are essential. Current Arabic text summarization systems face challenges such as language complexity and vocabulary limitations. We introduce an innovative framework using Arabic Named Entity Recognition to enhance abstractive summarization, crucial for NLP applications like question answering and knowledge graph construction. Our model, based on natural language generation techniques, adapts to diverse datasets. It identifies key information, synthesizes it into coherent summaries, and ensures grammatical accuracy through deep learning. Evaluated on the EASC dataset, our model achieved a 74% ROUGE1 score and a 97.6% accuracy in semantic coherence, with high readability and relevance scores. This sets a new standard for Arabic text summarization, greatly improving NLP information processing. © The Author(s) 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Adaptable and Dynamic Access Control Decision-Enforcement Approach Based on Multilayer Hybrid Deep Learning Techniques in BYOD Environment

引用

computers, Materials & Continua 2024年第9期80卷 4663-4686页

作者： Aljuaid Turkea Ayedh M Ainuddin Wahid Abdul Wahab Mohd Yamani Idna Idris Faculty of Computer Science and Information Technology Universiti MalayaKuala Lumpur50603Malaysia Faculty of Computer Science and Information Technology Shaqra UniversityShaqra11961Saudi Arabia Center for Mobile Cloud Computing Faculty of Computer Science and Information TechnologyUniversiti MalayaKuala Lumpur50603Malaysia

Organizations are adopting the Bring Your Own Device(BYOD)concept to enhance productivity and reduce ***,this trend introduces security challenges,such as unauthorized *** access control systems,such as Attribute-Based Access Control(ABAC)and Role-Based Access Control(RBAC),are limited in their ability to enforce access decisions due to the variability and dynamism of attributes related to users and *** paper proposes a method for enforcing access decisions that is adaptable and dynamic,based on multilayer hybrid deep learning techniques,particularly the Tabular Deep Neural Network Tabular DNN *** technique transforms all input attributes in an access request into a binary classification(allow or deny)using multiple layers,ensuring accurate and efficient access *** proposed solution was evaluated using the Kaggle Amazon access control policy dataset and demonstrated its effectiveness by achieving a 94%accuracy ***,the proposed solution enhances the implementation of access decisions based on a variety of resource and user attributes while ensuring privacy through indirect communication with the Policy Administration Point(PAP).This solution significantly improves the flexibility of access control systems,making themmore dynamic and adaptable to the evolving needs ofmodern ***,it offers a scalable approach to manage the complexities associated with the BYOD environment,providing a robust framework for secure and efficient access management.

关键词： BYOD security access control access control decision-enforcement deep learning neural network techniques TabularDNN multilayer dynamic adaptable flexibility bottlenecks performance policy conflict

来源：评论

学校读者我要写书评

暂无评论

Advances in neural architecture search

引用

National science Review 2024年第8期11卷 24-38页

作者： Xin Wang Wenwu Zhu Department of Computer Science and Technology Beijing National Research Center for Information Science and Technology Tsinghua University

Automated machine learning(AutoML) has achieved remarkable success in automating the non-trivial process of designing machine learning *** the focal areas of AutoML,neural architecture search(NAS) stands out,aiming to systematically explore the complex architecture space to discover the optimal neural architecture configurations without intensive manual *** has demonstrated its capability of dramatic performance improvement across a large number of real-world *** core components in NAS methodologies normally include(ⅰ) defining the appropriate search space,(ⅱ)designing the right search strategy and(ⅲ) developing the effective evaluation *** early NAS endeavors are characterized via groundbreaking architecture designs,the imposed exorbitant computational demands prompt a shift towards more efficient paradigms such as weight sharing and evaluation estimation,***,the introduction of specialized benchmarks has paved the way for standardized comparisons of NAS ***,the adaptability of NAS is evidenced by its capability of extending to diverse datasets,including graphs,tabular data and videos,etc.,each of which requires a tailored *** paper delves into the multifaceted aspects of NAS,elaborating on its recent advances,applications,tools,benchmarks and prospective research directions.

关键词： machine learning artificial intelligence neural architecture search

来源：评论

学校读者我要写书评

暂无评论

Robust steganographic approach using generative adversarial network and compressive autoencoder

引用

Multimedia Tools and Applications 2024年 1-38页

作者： Qasaimeh, Malik Qtaish, Alaa Abu Aljawarneh, Shadi Department of Computer Information Systems Faculty of Computer and Information Technology Jordan University of Science and Technology Irbid Jordan Department of Software Engineering co-joint with the Department of Cyber Security Faculty of Computer and Information Technology Jordan University of Science and Technology Irbid Jordan

Nowadays, social media applications and websites have become a crucial part of people’s lives;for sharing their moments, contacting their families and friends, or even for their jobs. However, the fact that these valuable data are transferred via the Internet and open channels, which are vulnerable to attacks and espionage, requires using defense methods to improve the security of data transmission. Recently, the progress of Deep Learning (DL) inspired researchers to use it with security methods such as steganography, the art of hiding secret data in unrelated content. DL and steganography combined enhanced the hiding properties, especially in the Coverless Image Steganography (CIS) methods. In this paper, we propose a Compressive Coverless Image Steganography (CCIS) model, which is a generated-based CIS, to improve the capacity and robustness of Steganography. This model uses the compressive autoencoder to compress the message, thus increasing the capacity, and uses the Generative Adversarial Network (GAN) to generate a stego image from the compressed vector, thus enhancing concealment ability, besides using a regression model and optimization method to extract the data. Another version was designed to enable sending binary messages by replacing the compressive autoencoder network with a robust mapping rule. Experiments show that the proposed model could improve the characteristics of Steganography. Furthermore, the proposed model could extract binary messages with 256 bits from attacked stego images with 99% recovery accuracy. Thus, capacity and robustness were enhanced using our model with both images and text as secret messages. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A new CNN-based watermarking method for color medical images in a fusion domain

引用

Neural Computing and Applications 2025年 1-19页

作者： Hosny, Khalid M. Abdel-Aziz, Mostafa M. Lashin, Nabil A. Hamza, Hanaa M. Department of Information Technology Faculty of Computers and Informatics Zagazig University Zagazig44519 Egypt Department of Information Technology Faculty of Information Technology and Computer Science Sinai University Arish Campus Al-Arish45518 Egypt

The transmission of medical images via medical agencies raises security concerns, necessitating increased security measures to ensure integrity and security. However, many watermarking algorithms overlook equipoise;the relation between robustness, invisibility, and payload capacity results in a less satisfactory performance. To bridge this gap, we propose a new blind watermarking method for securing medical images based on a convolutional neural network in a fusion domain. First, we transmute the host color image using a phase-only transform (PHOT) to detect the surface pattern. We feed the detected surface pattern into a pre-trained VGG19 model to impeccably extract the stable feature vector. We encrypt the extracted image features using a hybrid Chirikov map to enhance their robustness and reliability. The synergy between the pre-trained VGG19 model and PHOT in an integrated domain effectively captures additional native and localized image features. A hybrid Fibonacci Q-matrix method scrambles the binary watermark to enhance security. Integrating double encryption into the proposed method markedly enhances its resistance against diverse countermeasures. Experimental outcomes indicate that the proposed scheme performs efficiently in robustness and invisibility. The obtained PSNR value was up to 60.15 dB, which is optimal for human perception. The embedded watermark can be retrieved without any warping. The retrieved watermark seems to be authentic, exhibiting ideal BER and NC values. In almost all attack circumstances, the BER values approached zero while the NC values got closer. The proposed method demonstrates notable enhancements in robustness and invisibility compared to prior methods. © The Author(s) 2025.

关键词： Watermarking

来源：评论

学校读者我要写书评

暂无评论

Hybrid Deep Learning Approach for Automating App Review Classification:Advancing Usability Metrics Classification with an Aspect-Based Sentiment Analysis Framework

引用

computers, Materials & Continua 2025年第1期82卷 949-976页

作者： Nahed Alsaleh Reem Alnanih Nahed Alowidi Department of Computer Science Faculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddah21589Saudi Arabia Department of Computer Science College of Computer Science and EngineeringUniversity of HailHail81451Saudi Arabia

App reviews are crucial in influencing user decisions and providing essential feedback for developers to improve their *** the analysis of these reviews is vital for efficient review *** traditional machine learning(ML)models rely on basic word-based feature extraction,deep learning(DL)methods,enhanced with advanced word embeddings,have shown superior *** research introduces a novel aspectbased sentiment analysis(ABSA)framework to classify app reviews based on key non-functional requirements,focusing on usability factors:effectiveness,efficiency,and *** propose a hybrid DL model,combining BERT(Bidirectional Encoder Representations from Transformers)with BiLSTM(Bidirectional Long Short-Term Memory)and CNN(Convolutional Neural Networks)layers,to enhance classification *** analysis against state-of-the-art models demonstrates that our BERT-BiLSTM-CNN model achieves exceptional performance,with precision,recall,F1-score,and accuracy of 96%,87%,91%,and 94%,*** contributions of this work include a refined ABSA-based relabeling framework,the development of a highperformance classifier,and the comprehensive relabeling of the Instagram App Reviews *** advancements provide valuable insights for software developers to enhance usability and drive user-centric application development.

关键词： Requirements Engineering(RE) app review analysis usabilitymetrics hybrid deep learning BERT-BiLSTM-CNN

来源：评论

学校读者我要写书评

暂无评论

Malicious iOS Apps Detection Through Multi-Criteria Decision-Making Approach

引用

Informatica (Slovenia) 2025年第1期49卷 207-220页

作者： Bhatt, Arpita Jadhav Sardana, Neetu Department of Computer Science & Engineering and Information Technology Jaypee Institute of Information Technology Indonesia

In today’s era, smartphones are used in daily lives because they are ubiquitous and can be customized by installing third-party apps. As a result, the menaces because of these apps, which are potentially risky for user’s privacy, have increased. information on smartphones is perhaps, more personal than compared to data stored on desktops or computers, making it an easy target for intruders. After Android, the most prevalently used mobile operating system is Apple’s iOS. Both Android and iOS follow permission-based access control to protect user’s privacy. However, the users are unaware whether the app is breaching the user’s privacy. To combat this problem, in the paper we propose a hybrid approach to detect malicious iOS apps based on its permissions. In the first phase, weights have been assigned to app permissions using multi-criteria decision-making (MCDM) approach namely Analytic Hierarchy Process (AHP), and in the second phase machine learning& ensemble learning techniques have been employed to train the classifiers for detecting malicious apps. To test the efficacy of the proposed method dataset comprising 1150 apps from 12 app categories has been used. The results demonstrate the proposed approach improves the efficacy of detecting malicious iOS apps for majority of categories. © 2025 Slovene Society Informatika. All rights reserved.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Gender and age-evolution detection based on audio forensic analysis using light deep neural network

引用

International Journal of Speech technology 2023年第4期26卷 1091-1098页

作者： AL-Shakarchy, Noor D. Rageb, Huda Safoq, Mais Saad Department of Computer Science Faculty of Computer Science and Information Technology Kerbala University Kerbala Iraq

Forensic audio analysis is a foundation stone of many crime investigations. In forensic evidence;the audio file of the human voice is analyzed to extract much information in addition to the content of the speech, such as the speaker’s identity, emotions, gender, origin, etc. The accurate determination of individuals into groups based on their age development stage and their gender are often used as early investigations to differentiate them and determine the legal rights and responsibilities associated with them. This work introduces a light CNN model with a new architecture to detect the human age-evolution being’s stage (kids or adults) at the same time the gender of the adult one (male or female) based on the individual’s voice characteristics, which offers a balance between computational efficiency and model accuracy. The temporal information in the audio file is prepared by scaled and normalized. Then this information is exploited to extract and track the unique and salient audio features that make up the pattern of the feature map for each target class through some convolutional layers followed by maxpooling layers. Finally, The decision is made based on these feature maps by some fully connected layers. Successful and promising results are accomplished in terms of accuracy and loss functions which realize 0.99 and 0.017 respectively over the riched Voxceleb2 dataset. The proposed model underscores the importance of leveraging Light DNNs for gender and age-evolution detection, offering a robust and ethically sound solution for real-world applications in the field of audio forensics such as span speaker identification, victim profiling, deception detection, and more, contributing to the advancement of audio forensic analysis. © 2023, The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：