检索结果-内蒙古大学图书馆

4th International Conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Aishwarya, S. Selvamurugan, C. Parthiban, K.G. Nathan, B. Manoj Prabhakar, J. Mohamed Noordeen, A. Sri Eshwar college of Engineering Department of Computer Science and Engineering Coimbatore India Dhaanish Ahmed Institute of Technology Department of Science and Humanities Coimbatore India Dhaanish Ahmed Institute of Technology Department of Biomedical Engineering Coimbatore India Dhaanish Ahmed Institute of Technology Department of Computer Science and Engineering Coimbatore India Dhaanish Ahmed Institute of Technology Department of Artificial Intelligence & Data Science Coimbatore India

ISBN: (数字)9798331529635

ISBN: (纸本)9798331529635

GenAI has revolutionized the generation of realistic and imaginative data in ways that were previously beyond the capabilities of other machine learning algorithms. This area is rapidly gaining traction, with extensive research currently underway to advance it further. The most recent EY Reimagining Industry Futures Study highlights the generative AI (GenAI) to be a pivotal technology. The surge in GenAI's prominence can largely be attributed to advancements in deep neural networks, hardware technology such as GPUs and increased availability of data. This research study reviews the most popular GenAI models, spanning text, image, and audio modalities. This study provides a comparative analysis of various models, serving as a comprehensive resource for understanding the functionality and efficiency of the top GenAI models available today. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Performance analysis of message prioritization in delay tolerant networks

Performance analysis of message prioritization in delay tole...

引用

2012 IEEE Military Communications Conference, MILCOM 2012

作者： McAtee, Florence Narayanan, Sathya Xie, Geoffrey G. Computer Science and Information Technology Program California State University Monterey Bay Seaside CA 93955 United States Department of Computer Science Naval Postgraduate School Monterey CA 93943 United States

ISBN: (纸本)9781467317290

A variety of network deployments in disaster recovery, fire fighting and military scenarios create networks that do not form a connected network all the time. In such scenarios, paths between a source and destination may be created over time when nodes encounter each other due to node mobility. Many routing algorithms that can route messages in such delay-tolerant networking (DTN) settings have been proposed. Currently, these protocols do not differentiate between the messages that are being routed through the network treating all messages equally. But, there are DTN scenarios, in particular in military situations, where it is critical that messages are differentiated based on their importance to the task being supported. In this paper, we propose extensions to two popular DTN routing protocols, Epidemic and PROPHET, to implement such message prioritization. We implemented the proposed extensions in ONE (Opportunistic Network Emulator) and studied the performance of these extensions. Results from our simulation analysis demonstrate that it is possible to implement message prioritization in non-deterministic delay tolerant networks and provide us with insights into configuring message prioritization in DTN. In particular, the importance of selecting optimal queue lengths at each node to mitigate congestion in such networks. © 2012 IEEE.

关键词： Delay tolerant networks

来源：评论

学校读者我要写书评

暂无评论

Robust Perceptual Hashing for Fingerprint Template Protection 1

Robust Perceptual Hashing for Fingerprint Template Protectio...

引用

1st International Conference on computer Communications and Intelligent Systems, I3CIS 2022

作者： Birouk, Wafa Lahoulou, Atidel Melit, Ali Bouridane, Ahmed University of Bejaia Faculty of Exact Sciences Computer Science Department Bejaia06000 Algeria University Mohammed Seddik Benyahia Computer Science Department Jijel18000 Algeria LaRIA Laboratory University Mohammed Seddik Benyahia Computer Science Department Jijel18000 Algeria University of Sharjah Cybersecurity and Data Analytics Research Center Sharjah United Arab Emirates

ISBN: (纸本)9781665461887

Perceptual hashing has proven its robustness in securing biometric templates, where several methods have been proposed in order to have a robust and discriminating hash code against acceptable attacks. This paper proposes a robust perceptual hashing scheme for biometric template protection that based on SIFT and Harris functions for the feature ex-traction/selection phase and on DWT and SVD functions for the hash code generation phase. The experimental results in this paper discussed the effect of distance value and image blocks size parameters on hash performances (robustness and discrimination), which are used in the feature selection phase and the hash generation phase of our method respectively. The results show that we have could to choose the optimal parameters values in the generation of our hash code to give a best tradeoff between discrimination and robustness against acceptable image manipulations. © 2022 IEEE.

关键词： Signal reconstruction

来源：评论

学校读者我要写书评

暂无评论

SMOTE Oversampling and Near Miss Undersampling Based Diabetes Diagnosis from Imbalanced dataset with XAI Visualization 28

SMOTE Oversampling and Near Miss Undersampling Based Diabete...

引用

28th IEEE Symposium on computers and Communications, ISCC 2023

作者： Nayan, Nasim Mahmud Islam, Ashraful Islam, Muhammad Usama Ahmed, Eshtiak Hossain, Mohammad Mobarak Alam, Md Zahangir University of Information Technology and Sciences Department of Computer Science and Engineering Bangladesh Independent University Center for Computational and Data Sciences Bangladesh School of Computing and Informatics University of Louisiana at Lafayette United States Independent University Department of Computer Science and Engineering Bangladesh

ISBN: (纸本)9798350300482

This study investigated the predictive ability of ten different machine learning (ML) models for diabetes using a dataset that was not evenly distributed. Additionally, the study evaluated the effectiveness of two oversampling and undersampling methods, namely the Synthetic Minority Oversampling Technique (SMOTE) and the Near-Miss algorithm. Explainable Artificial Intelligence (XAI) techniques were employed to enhance the interpretability of the model's predictions. The results indicate that the extreme gradient boosting (XGB) model combined with SMOTE oversampling technique exhibited the highest accuracy and an F1-score of 99% and 1.00 respectively. Furthermore, the utilization of XAI methods increased the dependability of the model's decision-making process, rendering it more appropriate for clinical use. These results imply that integrating XAI with ML and oversampling techniques can enhance the early detection and management of diabetes, leading to better diagnosis and intervention. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Universally Instance-Optimal Mechanisms for Private Statistical Estimation 37

Universally Instance-Optimal Mechanisms for Private Statisti...

引用

37th Annual Conference on Learning Theory, COLT 2024

作者： Asi, Hilal Duchi, John C. Haque, Saminul Li, Zewei Ruan, Feng Apple United States Departments of Statistics and Electrical Engineering Stanford University United States Department of Computer Science Stanford University United States Department of Statistics and Data Science Northwestern University United States

We consider the problem of instance-optimal statistical estimation under the constraint of differential privacy where mechanisms must adapt to the difficulty of the input dataset. We prove a new instance specific lower bound using a new divergence and show it characterizes the local minimax optimal rates for private statistical estimation. We propose two new mechanisms that are universally instance-optimal for general estimation problems up to logarithmic factors. Our first mechanism, the total variation mechanism, builds on the exponential mechanism with stable approximations of the total variation distance, and is universally instance-optimal in the high privacy regime Ε ≤ 1/√n. Our second mechanism, the T-mechanism, is based on the T-estimator framework (Birgé, 2006) using the clipped log likelihood ratio as a stable test: it attains instance-optimal rates for any Ε ≤ 1 up to logarithmic factors. Finally, we study the implications of our results to robust statistical estimation, and show that our algorithms are universally optimal for this problem, characterizing the optimal minimax rates for robust statistical estimation. © 2024 H. Asi, J.C. Duchi, S. Haque, Z. Li & F. Ruan.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effects of Mutations on MUC1-C/ED Protein Stability and Antibody Binding: Structural Insight

引用

Biochemical and Biophysical Research Communications 2025年

作者： Dwipanjan Sanyal Danny Muzata Vladimir N. Uversky Surender Kharbanda Sourav Chowdhury Ravi Jasuja Xyone Therapeutics Boston Massachusetts USA Department of Computer Science and Information Systems Birla Institute of Technology and Science-Pilani Hyderabad India USF Health Byrd Alzheimerʼs Research Institute Morsani College of Medicine University of South Florida Tampa FL 33612 USA Department of Biological Sciences Birla Institute of Technology and Science-Pilani Hyderabad India Research Program in Men's Health: Aging and Metabolism Brigham and Women's Hospital Harvard Medical School Boston Massachusetts USA

Antibody-drug conjugates (ADCs) represent a transformative approach in cancer therapy, combining the specificity of monoclonal antibodies with targeted delivery of potent cytotoxic drugs to tumors. Discovery of novel antigens has been the hallmark for development of ADC therapeutics and MUC1 is one such oncoprotein, which has garnered renewed interest recently. The 3D1 antibody, engineered to bind specifically to the alpha 3 (α3) helix of the MUC1-C/extracellular domain, is being actively developed towards clinical translation. The study examined the conformational coupling in the residues spanning the MUC1 sea urchin sperm protein, enterokinase, and agrin domain (SEA), which incorporates MUC1-C. 3D1 has been shown to be sensitive to three specific point mutations—D85E, V86A, and T88A— at the interaction surface between MUC1-C and the 3D1 antibody. Our findings reveal that D85E and T88A mutations cause significant conformational shifts and reduced binding affinity due to altered electrostatic interactions and hydrogen bonding. The V86A mutation, while maintaining the overall conformation, disrupts local hydrophobic interactions, leading to decreased binding efficiency. Further, residue 85 in the wild-type (WT) MUC1 SEA domain interacts with other residues, contributing to local frustration and destabilization. Mutants D85E, V86A, and T88A exhibited similar frustration patterns but with increased local energy, indicating reduced stability. Frustration calculations highlighted the crucial role of the α3 helix in protein stability and antibody interaction. Collectively, these data highlight the importance of specific residues contributing to the local frustration in maintaining effective antibody binding and suggest that mutations impacting these residues can significantly alter antigen binding dynamics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SWARM: Sensor driven water and agricultural resources management

SWARM: Sensor driven water and agricultural resources manage...

引用

2006 Annual India Conference, INDICON

作者： Babu, Chitra Athmaraman, Naren Ganesan, Archana Soundararajan, Srivathsan Muthukumar, Rama IEEE Department of Computer Science and Engineering SSN College of Engineering Chennai India Department of Computer Science University of Illinois at Chicago Chicago IL United States Department of Computer Science University of Texas at Austin Austin TX United States Department of Electrical Engineering Auburn University Auburn AL United States Verizon Data Services India Ltd. Chennai India

ISBN: (纸本)1424403707

Efficiently managing the use of water and fertilizers has always been a major concern in agricultural practices. The scarcity of fresh water has been one of the motivating factors in developing efficient irrigation systems. Apart from the environmental concern in water management, it facilitates better growth of the crops. Fertilizers in low quantities can considerably retard crop growth and hence affect production. Synthetic fertilizers in high quantities pollute the soil and render ground water unsuitable for consumption. Measuring the nutrient availability in real time and properly delivering the right amount of the nutrients at the appropriate time can be immensely useful. The primary objective of SWARM is to remotely monitor the water and nutrient availability and deliver the corresponding resources dynamically in accordance with the data collected. The SWARM system was implemented at a farm measuring 250 ft × 80 ft in Kuppam, Andhra Pradesh, India. Groundnut is being cultivated in the farm. It was tested for a period of around one and a half months and results have been provided. The system has shown promising results when compared to manually supplying water and fertilizers. The SWARM software was written in the JAVA programming language. @w006 IEEE.

关键词： Fertilizers

来源：评论

学校读者我要写书评

暂无评论

Automatical Pulmonary Nodule Detection by Feature Contrast Learning 1

引用

15th International Conference on Intelligent Computing, ICIC 2019

作者： Chang, Jie Ye, Minquan Gu, Naijie Zhang, Xiaoci Lin, Chuanwen Ye, Hong School of Computer Science and Technology University of Science and Technology of China Hefei230000 China Research Center of Health Big Data Mining and Applications Wannan Medical College WuhuAnhui241002 China Department of Computer Science and Technology Hefei University Hefei230601 China

ISBN: (数字)9783030267636

ISBN: (纸本)9783030267629

With regard to pulmonary nodule detection, due to the similar texture and shape as particular tissues, it is difficult for computer-Aided Detection (CAD) system in detecting pulmonary nodule with both high accuracy and sensitivity. To address this problem, we design a 3D automated pulmonary nodule detection where a auxiliary 3D generative adversarial network is embedded. This well-trained auxiliary component that fully learns volumetrically contextual information of nodule and non-nodule structure, is exploited for each input sample of detection model to generate a derivative which only preserve background context by removing all the nodules. By learning the feature contrast between each input and its derivative, our detection model achieves competitive performance to state-of-the-art approaches for the pulmonary nodule detection task. © 2019, Springer Nature Switzerland AG.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

A fast two-stage algorithm for computing SimRank and its extensions

A fast two-stage algorithm for computing SimRank and its ext...

引用

1st International Workshop on Graph database, IWGD 2010, 1st International Workshop on Advanced Techniques on XML data Management, XMLDM 2010, 2nd International Workshop on WCMT 2010, Held in Conj. with WAIM 2010

作者： Jia, Xu Liu, Hongyan Zou, Li He, Jun Du, Xiaoyong Key Labs. of Data Engineering and Knowledge Engineering Ministry of Education China Department of Computer Science Renmin University of China 100872 China Department of Management Science and Engineering Tsinghua University 100084 China

ISBN: (纸本)3642167195

Similarity estimation can be used in many applications such as recommender system, cluster analysis, information retrieval and link prediction. SimRank is a famous algorithm to measure objects' similarities based on link structure. We observe that if one node has no in-link, similarity score between this node and any of the others is always zero. Based on this observation, we propose a new algorithm, fast two-stage SimRank (F2S-SimRank), which can avoid storing unnecessary zeros and can accelerate the computation without accuracy loss. Under the circumstance of no accuracy loss, this algorithm uses less computation time and occupies less main memory. Experiments conducted on real and synthetic datasets demonstrate the effectiveness and efficiency of our F2S-SimRank. © 2010 Springer-Verlag.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Interpretability adversarial example detection through multi-interpretation aggregation comparison

引用

Applied Soft Computing 2025年 177卷

作者： Chen, Zigang Wang, Zhangqi Pan, Ding Leng, Tao Liu, Yuhong Zhu, Haihua Chongqing Key Laboratory of Cyberspace Security Monitoring and Governance Chongqing University of Posts and Telecommunications No. 2 Chongwen Road Nan'an District Chongqing 400065 China Intelligent Policing Key Laboratory of Sichuan Province Sichuan Police College Sichuan 646000 China Department of Computer Science and Engineering Santa Clara University 500 El Camino Real Santa ClaraCA95053 United States Key Laboratory of Cyberspace Big Data Intelligent Security Ministry of Education Chongqing University of Posts and Telecommunications No. 2 Chongwen Road Nan'an District Chongqing 400065 China

Deep neural network adversarial attacks have extended to the domain of interpretability, resulting in interpretation manipulation. Therefore, we propose a novel approach for detecting explanatory adversarial examples based on multi-interpretation aggregation comparison, called AGGEC (Aggregate Explain Compare). Our approach employs the aggregation of multiple different interpretation results to enable a comprehensive analysis of the disparities in texture shapes before and after the interpretation aggregation and employs gray level co-occurrence matrix to extract texture features before and after the interpretation aggregation to accentuate the discernible distinctions. These dissimilarities are subsequently utilized to train an external detector. AGGEC exhibits exceptional detection performance in black-box, gray-box, and white-box scenarios, achieving detection success rates of 99.4% and 95.6% on CIFAR-10 and ImageNet, respectively. Empirical results substantiate the efficacy of our proposed method in effectively identifying malicious manipulation of interpretation results. © 2025 Elsevier B.V.

关键词： Deep learning

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：