检索结果-内蒙古大学图书馆

Secure and Trustworthy Machine Learning (SaTML), IEEE Conference on

作者： Daryna Oliynyk Rudolf Mayer Andreas Rauber CDL AsTra Faculty of Computer Science University of Vienna Vienna Austria SBA Research Vienna Austria Faculty of Informatics TU Wien Institute of Information Systems Engineering Vienna Austria

ISBN: (数字)9798331517113

ISBN: (纸本)9798331517120

Machine learning (ML) models were shown to be vulnerable to model stealing attacks, which lead to intellectual property infringement. Among other attack methods, substitute model training is an all-encompassing attack applicable to any machine learning model whose behaviour can be approximated from input-output queries. Whereas previous works mainly focused on improving the performance of substitute models by, e.g. developing a new substitute training method, there have been only limited ablation studies that try to understand the impact the strength of an attacker has on the substitute model's performance. As a result, different authors came to diverse, sometimes contradicting, conclusions. In this work, we exhaustively examine the ambivalent influence of different factors resulting from varying the attacker's capabilities and knowledge on a substitute training attack. Our findings suggest that some of the factors that have been considered important in the past are, in fact, not that influential; instead, we discover new correlations between attack conditions and success rate. In particular, we demonstrate that better-performing target models enable higher-fidelity attacks and explain the intuition behind this phenomenon. Further, we propose to shift the focus from the complexity of target models toward the complexity of their learning tasks. Therefore, for the substitute model, rather than aiming for a higher architecture complexity, we suggest focusing on getting data of higher complexity and an appropriate architecture. Finally, we demonstrate that even in the most limited data-free scenario, there is no need to overcompensate weak knowledge with unrealistic capabilities in the form of millions of queries. Our results often exceed or match the performance of previous attacks that assume a stronger attacker, suggesting that these stronger attacks are likely endangering a model owner's intellectual property to a significantly higher degree than shown until now.

关键词： Training Correlation Query processing Focusing Intellectual property Data collection Data models Adversarial machine learning Complexity theory Tuning

来源：评论

学校读者我要写书评

暂无评论

A comparative study between android phone and TV apps

引用

Automated Software Engineering 2025年第2期32卷 1-30页

作者： Liu, Yonghui Chen, Xiao Liu, Yue Kong, Pingfan Bissyandé, Tegawendé F. Klein, Jacques Sun, Xiaoyu Li, Li Chen, Chunyang Grundy, John Department of Software Systems and Cybersecurity Monash University Melbourne Australia School of Information and Physical Sciences University of Newcastle Newcastle Australia SnT University of Luxembourg Esch-sur-Alzette Luxembourg The School of Computing Australian National University Canberra Australia School of Software Beihang University Beijing China Department of Computer Science Technical University of Munich Munich Germany Faculty of Data Science City University of Macau China

Smart TVs have surged in popularity, leading developers to create TV versions of mobile apps. Understanding the relationship between TV and mobile apps is key to building consistent, secure, and optimized cross-platform experiences while addressing TV-specific SDK challenges. Despite extensive research on mobile apps, TV apps have been given little attention, leaving the relationship between phone and TV apps unexplored. Our study addresses this gap by compiling an extensive collection of 3445 Android phone/TV app pairs from the Google Play Store, launching the first comparative analysis of its kind. We examined these pairs across multiple dimensions, including non-code elements, code structure, security, and privacy aspects. Our findings reveal that while these app pairs could get identified with the same package names, they deploy different artifacts with varying functionality across platforms. TV apps generally exhibit less complexity in terms of hardware-dependent features and code volume but maintain significant shared resource files and components with their phone versions. Interestingly, some categories of TV apps show similar or even severe security and privacy concerns compared to their mobile counterparts. This research aims to assist developers and researchers in understanding phone-TV app relationships, highlight domain-specific concerns necessitating TV-specific tools, and provide insights for migrating apps from mobile to TV platforms. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Television applications

来源：评论

学校读者我要写书评

暂无评论

Multiscale Frequency-Guided Image Analyses for Mixed-Modality Medical Image Segmentation

引用

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Lin, Zijin Huang, Guoheng Chen, Zhuoheng Li, Yan Yuan, Xiaochen Pun, Chi-Man Cheng, Lianglun Guangdong University of Technology School of Computer Science and Technology Guangzhou510006 China Shenzhen Polytechnic University School of Artificial Intelligence Shenzhen518055 China Macao Polytechnic University Faculty of Applied Sciences China University of Macau Department of Computer and Information Science China

Medical images generated by different imaging instruments often have different modalities. An architecture that can bridge the instrument gap and provide unified training for different modalities of images is the development direction of medical image segmentation. However, existing methods seldom pay attention to the mutual interference caused by the distribution differences of features between different modalities, which is a key factor in the training of mixed-modality datasets. Differences in diseases and imaging methods result in variations in the distribution of lesion size and frequency information. When training with data from different modalities mixed together, these differences typically lead to a decrease in network performance. To address this, we propose a mixed-modality segmentation network (MMSNet), which consists of three key components: multiscale frequency guidance (MSFG), modality feature adaptor (MFA), and frequency enhancement prompt (FEP). The MSFG refines the process of spatial feature extraction by incorporating multiscale frequency features. Moreover, MFA is designed to adjust pre-trained patch embeddings and transformer layers to reduce the cost of acquiring unknown modality features. Finally, FEP captures multiresolution frequency features of the image and fuses them with the frequency feature map of the image, which can enhance the network’s ability to extract frequency information across multiple spatial domains to a certain extent. In addition, we also introduced Mix1 and Mix2, which are composed of medical images from four different modalities to test the segmentation performance of MMSNet. Our experiment demonstrates that MMSNet can effectively alleviate the interference caused by differences in image frequency distribution, ultimately improving the segmentation quality of medical images. © 1963-2012 IEEE.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Certificate-Less Healthcare Signature Scheme for Secure Consumer Technology-Centric Wireless Medical Sensor Devices

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Vimal, Vrince Raina, Vikas Alghamdi, Azzah Majumdar, Parijata Shabaz, Mohammad Rao, G Siva Nageswara Nimma, Divya Keshta, Ismail Soni, Mukesh Mody University of Science and technology Department of CSE SET Laxmangarh Rajasthan SikarNH-52 India Imam Abdalrhman Bin Faisal University College of Computer Science and Information Technology Computer Information Systems Department Khobar Saudi Arabia Indian Institute of Information Technology Department of Computer Science and Engineering Agartala India Model Institute of Engineering and Technology J and K Jammu India Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering AP Vaddeswaram India The University of Southern Mississippi United States AlMaarefa University College of Applied Sciences Computer Science and Information Systems Department Riyadh Saudi Arabia Lovely Professional University Division of Research and Development Phagwara India

The application of Consumer-Centric Wireless Medical Sensor Devices in the medical field has brought great convenience to doctors and patients. However, the transmission process may compromise patient privacy, making the authenticity and integrity of medical data a significant challenge in the healthcare industry. To address this issue, a certificate-less Healthcare Signature scheme is proposed to protect patient privacy. However, most schemes are unable to resist fully selective cipher text attacks and suffer from key exposure problems. To mitigate these issues, a scheme is proposed to resist fully selective attacks in real-time processing of Consumer-Centric Wireless Medical Sensor Devices with high latency issues, while utilizing parallel key isolation methods to address key exposure problems. The scheme's enforceability is proven based on the random oracle model and the computational Diffie-Hellman assumption, and its resistance to fully selective key attacks is proven based on hash-based collision resistance. It is shown that this plan works well for wireless sensor networks in the medical field. Sensor nodes have limited resources, and sending vast volumes of patient medical data online significantly raises processing and communication costs. The low processing power of edge devices limits the effectiveness and performance of medical services. Efficiency analysis using the MIRACL library demonstrates that the proposed scheme outperforms other relevant schemes in terms of efficiency at each stage, while also having lower communication overhead. © 1975-2011 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

DFE: Deep Flow Embedding for Robust Network Traffic Classification

引用

IEEE Transactions on Network science and Engineering 2025年第3期12卷 1597-1612页

作者： Wang, Zhijiong Zhang, Anguo Li, Hung Chun Yin, Yadong Chen, Wei Lam, Chan Tong Mak, Peng Un Vai, Mang I. Gao, Yueming Pun, Sio Hang Zhuhai UM Science and Technology Research Institute Lingyange Semiconductor Incorporated Joint Laboratory Zhuhai519031 China Lingyange Semiconductor Inc. Zhuhai519030 China University of Macau State Key Laboratory of Analog and Mixed-Signal VLSI Institute of Microelectronics 999078 China Fuzhou University School of Physics and Information Engineering Fuzhou350108 China Fudan University Center for Intelligent Medical Electronics School of Information Science and Technology Human Phenome Institute Shanghai200437 China Fudan University Human Phenome Institute Shanghai200437 China Macau Polytechnic University Faculty of Applied Sciences 999078 China University of Macau Department of Electrical and Computer Engineering Faculty of Science and Technology 999078 China Fuzhou University Key Lab of Medical Instrumentation and Pharmaceutical Technology of Fujian Province Fuzhou350108 China

People's increasing demand for high-quality network services has prompted the continuous attention and development of network traffic classification (NTC). In recent years, deep flow inspection (DFI) is considered to be the most effective and promising method to solve the NTC. However, DFI still cannot effectively address the problem of changes in flow characteristics of complex packet flows and the discovery of new traffic categories. In this paper, we propose a metric learning based deep learning solution with feature compressor, named deep flow embedding (DFE). The feature compressor is used to compress the feature information transmitted layer by layer in DL backbone while maintaining the computational accuracy, so that the backbone can remove as much noise, redundancy, and other irrelevant information from the input data as possible, and achieve more robust feature extraction of network traffic flow. The deep learning (DL) backbone generates an embedding vector for each network packet flow. Then the embedding vector is compared with the vector template preset for each traffic type in the template library to determine the category of the packet flow. Experimental results verify that our method is more effective than the traditional DFI methods in overcoming the problems of flow characteristics variation and new category discovery. © 2013 IEEE.

关键词： Network embeddings

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph-based Retrieval-Augmented Generation for Schema Matching

arXiv

引用

arXiv 2025年

作者： Ma, Chuangtao Chakrabarti, Sriom Khan, Arijit Molnár, Bálint The Department of Computer Science Aalborg University Aalborg Denmark The Department of Information Systems Faculty of Informatics Eötvös Loránd University Budapest Hungary

Traditional similarity-based schema matching methods are incapable of resolving semantic ambiguities and conflicts in domain-specific complex mapping scenarios due to missing commonsense and domain-specific knowledge. The hallucination problem of large language models (LLMs) also makes it challenging for LLM-based schema matching to address the above issues. Therefore, we propose a Knowledge Graph-based Retrieval-Augmented Generation model for Schema Matching, referred to as the KG-RAG4SM. In particular, KG-RAG4SM introduces novel vector-based, graph traversal-based, and query-based graph retrievals, as well as a hybrid approach and ranking schemes that identify the most relevant subgraphs from external large knowledge graphs (KGs). We showcase that KG-based retrieval-augmented LLMs are capable of generating more accurate results for complex matching cases without any re-training. Our experimental results show that KG-RAG4SM outperforms the LLM-based state-of-the-art (SOTA) methods (e.g., Jellyfish-8B) by 35.89% and 30.50% in terms of precision and F1 score on the MIMIC dataset, respectively;KG-RAG4SM with GPT-4o-mini outperforms the pre-trained language model (PLM)-based SOTA methods (e.g., SMAT) by 69.20% and 21.97% in terms of precision and F1 score on the Synthea dataset, respectively. The results also demonstrate that our approach is more efficient in end-to-end schema matching, and scales to retrieve from large KGs. Our case studies on the dataset from the real-world schema matching scenario exhibit that the hallucination problem of LLMs for schema matching is well mitigated by our solution. Copyright © 2025, The Authors. All rights reserved.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Predicting Concentration Levels of Air Pollutants by Transfer Learning and Recurrent Neural Network

arXiv

引用

arXiv 2025年

作者： Fong, Iat Hang Li, Tengyue Fong, Simon Wong, Raymond K. Tallón-Ballesteros, Antonio J. Department of Computer and Information Science University of Macau Taipa China Faculty of Engineering University of New South Wales Sydney Australia Department of Languages and Computer Systems University of Seville Seville Spain

Air pollution (AP) poses a great threat to human health, and people are paying more attention than ever to its prediction. Accurate prediction of AP helps people to plan for their outdoor activities and aids protecting human health. In this paper, long-short term memory (LSTM) recurrent neural networks (RNNs) have been used to predict the future concentration of air pollutants (APS) in Macau. Additionally, meteorological data and data on the concentration of APS have been utilized. Moreover, in Macau, some air quality monitoring stations (AQMSs) have less observed data in quantity, and, at the same time, some AQMSs recorded less observed data of certain types of APS. Therefore, the transfer learning and pre-trained neural networks have been employed to assist AQMSs with less observed data to build a neural network with high prediction accuracy. The experimental sample covers a period longer than 12-year and includes daily measurements from several APS as well as other more classical meteorological values. Records from five stations, four out of them are AQMSs and the remaining one is an automatic weather station, have been prepared from the aforesaid period and eventually underwent to computational intelligence techniques to build and extract a prediction knowledge-based system. As shown by experimentation, LSTM RNNs initialized with transfer learning methods have higher prediction accuracy;it incurred shorter training time than randomly initialized recurrent neural networks. Copyright © 2025, The Authors. All rights reserved.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

EnVoR: Multi-Granular Gold Price Prediction Using Ensemble Voting Regression

EnVoR: Multi-Granular Gold Price Prediction Using Ensemble V...

引用

Artificial Intelligence, Applications, Innovation and Ethics (AI2E), International Conference for

作者： Rajkumar Kannan Samuel Giftson Durai Abdul Raouf Khan Kalaiarasi Sonai Muthu Department of Computer Science Bishop Heber College (Autonomous) Tiruchirappalli India Department of Information Technology University of Technology and Applied Sciences Muscat Oman Department of Computer Sciences King Faisal University Alhassa Saudi Arabia Faculty of Information Science and Technology Multimedia University Malaysia

ISBN: (数字)9798331513733

ISBN: (纸本)9798331513740

Price prediction and forecasting have been important topics in the financial world. Among all commodities, such as gold, silver, platinum and others, gold has been the prime and forefront commodity which decides the price movements of other commodities. This paper proposes a unique multi-granular gold price prediction methodology, called EnVoR, by leveraging voting based regression tree models and utilizing simple, cumulative and exponential moving average values as features. Our proposed EnVoR predicts gold prices of next day, next week, next month and next quarter seasons. Extensive experiments are conducted on the gold price daily data from December 2011 to June 2024 to predict the future prices at various granularity levels. Experimental results on the gold price data have confirmed that the proposed EnVoR methodology has given superior performance in terms of highest regression accuracy and the lowest error values for MSE, RMSE, MAE and MAPE for predicting next day, next week, next month and next quarter price values. The proposed EnVoR model can be used by financial institutions and stock market analysts for providing buy, sell or hold recommendations to their clients.

关键词： Gold Silver Technological innovation Tracking Biological system modeling Market research Vectors Random forests Regression tree analysis Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Empowering Dysarthric Communication: Hybrid Transformer-CTC-Based Speech Recognition System

引用

IEEE Access 2025年 13卷 82479-82491页

作者： Vinotha, R. Hepsiba, D. Anand, L. D. Vijay Bruntha, P. Malin Dinh, Linh Dang, Hien Karunya Institute of Technology and Sciences Department of Robotics Engineering Tamil Nadu Coimbatore641114 India Karunya Institute of Technology and Sciences Department of Biomedical Engineering Tamil Nadu Coimbatore641114 India Karunya Institute of Technology and Sciences Department of Electronics and Communication Engineering Tamil Nadu Coimbatore641114 India Suffolk University Department of Information Systems BostonMA02108 United States Thuyloi University Faculty of Computer Science and Engineering Hanoi116705 Viet Nam Molloy University Department of Mathematics and Computer Science Rockville CentreNY11570 United States

Dysarthria, a motor speech disorder, impairs the muscles involved in speech production, leading to challenges in articulation, pronunciation, and overall communication. This results in slow, slurred speech that is difficult to understand. Augmentative and Alternative Communication (AAC) aids integrated with speech recognition technology offer a promising solution for individuals with dysarthria. However, Automatic Speech Recognition (ASR) systems trained on typical speech data often struggle to recognize dysarthric speech due to its unique speech patterns and limited training data. To address these challenges, a hybrid Transformer-CTC model has been proposed for improving ASR performance on dysarthric speech. The Transformer architecture employs a self-attention mechanism that models complex dependencies between speech features, enabling it to identify and emphasize important patterns even when training data is limited. This ability is particularly crucial for dysarthric speech, where speech signals often exhibit high variability. On the other hand, Connectionist Temporal Classification (CTC) acts as an effective transcription layer. It aligns speech features with character sequences without requiring precise input-output alignment, making it well-suited for handling the inconsistencies and distortions present in dysarthric speech. The integration of these components creates a powerful architecture capable of learning nuanced speech patterns and delivering accurate transcriptions for dysarthric speech. The model was trained using the UA speech corpus, containing 13 hours of speech from 15 speakers with varying dysarthria levels. The proposed hybrid system achieves an impressive Word Recognition Accuracy (WRA) of 89%, demonstrating its effectiveness in accurately transcribing dysarthric speech. This innovative approach significantly advances the development of ASR technologies tailored to diverse and variable speech patterns, ultimately enhancing communication for in

关键词： Speech analysis

来源：评论

学校读者我要写书评

暂无评论

Machine Learning and Optimization Approaches for Dissolved Oxygen Prediction in River systems

Machine Learning and Optimization Approaches for Dissolved O...

引用

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Hazem Alnajjar Nadia Al-Rousan Douaa Al-Rouan Department of Computer Science College of Information Technology University of Petra Amman Jordan Department of computer engineering College of Engineering AlBalqa Applied University Salt Jordan Department of Computer Engineering Faculty of Electrical Engineering and Information Systems German Jordanian University Amman Jordan School of Civil Engineering Universiti Sains Malaysia Nibong Tebal Pulau Pinang MALAYSIA

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

This study aims to combine optimization algorithms of feature selection and machine learning models for DO prediction at a water monitoring station. The dataset consisted of water quality indicators from three stations, whose DO was the target variable at the third station. This study used seven optimization algorithms to select the most relevant features. The selected features were then used along with ten machine-learning models for building predictive models. Several models were evaluated based on R 2 , MAE, and MSE metrics. Hence, Gradient Boosting and Random Forest models yielded better prediction results. The results showed high accuracy, reflected by the values of R 2 , which were 0.987 and 0.922 for the training data and 0.769 and 0.745 for the test data, respectively. The findings underscore the effectiveness of combining feature selection with optimized machine learning models for accurate DO prediction, which is essential for monitoring water quality.

关键词： Machine learning algorithms Biological system modeling Water quality Predictive models Prediction algorithms Feature extraction Rivers Water monitoring Optimization Random forests

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：