检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Murali, Nihal Puli, Aahlad Yu, Ke Ranganath, Rajesh Batmanghelich, Kayhan Intelligent Systems Program University of Pittsburgh United States Department of Computer Science New York University United States Department of Electrical and Computer Engineering Boston University United States

Deep Neural Networks (DNNs) are prone to learning spurious features that correlate with the label during training but are irrelevant to the learning problem. This hurts model generalization and poses problems when deploying them in safety-critical applications. This paper aims to better understand the effects of spurious features through the lens of the learning dynamics of the internal neurons during the training process. We make the following observations: (1) While previous works highlight the harmful effects of spurious features on the generalization ability of DNNs, we emphasize that not all spurious features are harmful. Spurious features can be "benign" or "harmful" depending on whether they are "harder" or "easier" to learn than the core features for a given model. This definition is model and dataset dependent. (2) We build upon this premise and use instance difficulty methods (like Prediction Depth (Baldock et al., 2021)) to quantify "easiness" for a given model and to identify this behavior during the training phase. (3) We empirically show that the harmful spurious features can be detected by observing the learning dynamics of the DNN’s early layers. In other words, easy features learned by the initial layers of a DNN early during the training can (potentially) hurt model generalization. We verify our claims on medical and vision datasets, both simulated and real, and justify the empirical success of our hypothesis by showing the theoretical connections between Prediction Depth and information-theoretic concepts like V-usable information (Ethayarajh et al., 2021). Lastly, our experiments show that monitoring only accuracy during training (as is common in machine learning pipelines) is insufficient to detect spurious features. We, therefore, highlight the need for monitoring early training dynamics using suitable instance difficulty metrics. © 2023, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Empowering Urban Accessibility: A Prototype System for People with Disability

Empowering Urban Accessibility: A Prototype System for Peopl...

引用

International Conference on Advanced Learning Technologies (ICALT)

作者： Arlley Dos Santos Costa Luis Carlos Costa Fonseca Sofiane Labidi Graduate Program in Computer and Systems Engineering State University of Maranhão – UEMA São Luís Brazil Bachelor of Science and Technology Program -BICT Federal University of Maranhão – UFMA São Luis Brazil

ISBN: (数字)9798350362053

ISBN: (纸本)9798350362060

This work demonstrates the development of a prototype system that could improve urban and digital accessibility for People with Disability - PwD. The research involves using an academic questionnaire to enable a wider perspective of the central subject and assessing the effectiveness of the proposed solution. The main goal is to help social integration and make the future solution implementable for smart cities through an experiment in an academic campus. The dissertation is related to technological and social issues that might be used as a resource by researchers, public or private bodies interested in more profound studies. The document recognizes the lack of studies on this subject, especially in Brazil because there are few available articles. The prototype system is designed to tackle the issues faced by persons with disability in urban settings through platforms, applications and mobile phones as it proposes a comprehensive set of solutions when addressing accessibility for PwDs. In turn, this research fosters knowledge advancements in the field of urban accessibility and provides vital recommendations for developing inclusive technologies and policies.

关键词： Smart cities Prototypes People with disabilities Mobile handsets

来源：评论

学校读者我要写书评

暂无评论

CLINICAL DETERIORATION PREDICTION IN BRAZILIAN HOSPITALS BASED ON ARTIFICIAL NEURAL NETWORKS AND TREE DECISION MODELS 15

CLINICAL DETERIORATION PREDICTION IN BRAZILIAN HOSPITALS BAS...

引用

15th International Conference on ICT, Society and Human Beings, ICT 2022, 19th International Conference on Web Based Communities and Social Media, WBC 2022 and 14th International Conference on e-Health, EH 2022 - Held at the 16th Multi Conference on computer science and Information systems, MCCSIS 2022

作者： Yazdanpanah, Hamed Silva, Augusto C.M. Guedes, Murilo Morales, Hugo M.P. dos Coelho, Leandro S. Moro, Fernando G. Department of Research Robô Laura PR Curitiba Brazil Department of Computer Science University of São Paulo SP São Paulo Brazil Industrial and Systems Engineering Graduate Program Pontifical Catholic University of Parana PR Curitiba Brazil Electrical Engineering Graduate Program Federal University of Parana PR Curitiba Brazil

ISBN: (纸本)9789898704405

Early recognition of clinical deterioration (CD) has vital importance in patients’ survival from exacerbation or death. Electronic health records (EHRs) data have been widely employed in Early Warning Scores (EWS) to measure CD risk in hospitalized patients. Recently, EHRs data have been utilized in Machine Learning (ML) models to predict mortality and CD. The ML models have shown superior performance in CD prediction compared to EWS. Since EHRs data are structured and tabular, conventional ML models are generally applied to them, and less effort is put into evaluating the artificial neural network’s performance on EHRs data. Thus, in this article, an extremely boosted neural network (XBNet) is used to predict CD, and its performance is compared to eXtreme Gradient Boosting (XGBoost) and random forest (RF) models. For this purpose, 103,105 samples from thirteen Brazilian hospitals are used to generate the models. Moreover, the principal component analysis (PCA) is employed to verify whether it can improve the adopted models’ performance. The performance of ML models and Modified Early Warning Score (MEWS), an EWS candidate, are evaluated in CD prediction regarding the accuracy, precision, recall, F1-score, and geometric mean (G-mean) metrics in a 10-fold cross-validation approach. According to the experiments, the XGBoost model obtained the best results in predicting CD among Brazilian hospitals’ data. © 2022 15th International Conference on ICT, Society and Human Beings, ICT 2022, 19th International Conference on Web Based Communities and Social Media, WBC 2022 and 14th International Conference on e-Health, EH 2022 - Held at the 16th Multi Conference on computer science and Information systems, MCCSIS 2022. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

A Comparison Diagnosis Algorithm for Conditional Fault Local Diagnosis of Multiprocessor systems 25th

A Comparison Diagnosis Algorithm for Conditional Fault Loca...

引用

25th International computer Symposium on New Trends in computer Technologies and Applications, ICS 2022

作者： Lv, Yali Lin, Cheng-Kuan Hsu, D. Frank Fan, Jianxi School of Information Technology Henan University of Chinese Medicine Zhengzhou China Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan Undergraduate Degree Program of Systems Engineering and Technology National Yang Ming Chiao Tung University Hsinchu Taiwan Computer Science and Information Engineering Chung Cheng Institute of Technology NDU Taoyuan City Taiwan Department of Computer and Information Science Fordham University New York United States School of Computer Science and Technology Soochow University Suzhou China

ISBN: (纸本)9789811995811

An efficient diagnosis is very important for a multiprocessor system. In this paper, we present a (α, β) -trees combination S(u, X, α, β) and give some conclusions about the local diagnosis. Based on the (α, β) -trees combination, we give a conditional fault local diagnosis algorithm to identify the fault or fault-free status of each processor correctly under the MM ∗ model when the number of faulty nodes does not exceed α+ 2 β- 3 and every node has at least one fault-free neighboring node. According to our results, a connected network with a (α, β) -trees combination S(u, X, α, β) for a node u is conditionally locally (α+ 2 β- 3 ) -diagnosable at node u and the time complexity of our algorithm to diagnose u is O(α2β+ αβ2). As an application, we show that our algorithm can identify all the faulty nodes of n-dimensional star graph Sn if the faulty node number does not exceed 3 n- 8. Compared with existing algorithms, our algorithm allows more faulty node in a multiprocessor system. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Computational complexity

来源：评论

学校读者我要写书评

暂无评论

Model for Anomaly Detection in Reducing Water Losses in the Distribution Network of Peru using Machine Learning Algorithms

Model for Anomaly Detection in Reducing Water Losses in the ...

引用

Software engineering and computer science (CSECS), International Conference on

作者： Mayra Rivera Pricila Lucero José Luis Castillo-Sequera Lenis Wong Information Systems Engineering Program Universidad Peruana de Ciencias Aplicadas Lima Peru Department of Computer Science Group: Health Computing and Intelligent System Universidad de Alcalá Alcalá de Henares Spain

ISBN: (数字)9798331522216

ISBN: (纸本)9798331522223

Water leakage in distribution networks is a significant challenge, especially in regions with limited infrastructure like Huancayo, Peru, where losses account for 32.82% of the distributed volume. This study introduces a machine learning-based approach to detect leaks using four algorithms: Autoencoder LSTM, Isolation Forest, One-Class SVM, and K-Nearest Neighbors (KNN). The methodology involved preprocessing historical consumption data (2018–2024) into 12-month temporal sequences per client and evaluating the models based on F1 Score, Precision, and Mean Absolute Error (MAE). Among the algorithms, the Autoencoder LSTM demonstrated superior performance with the highest precision (0.89) and the lowest MAE (0.00402). Its robustness in high-variability contexts enables early and reliable leak detection, providing a cost-effective solution for optimizing water management in resource-constrained environments.

关键词： Support vector machines Machine learning algorithms Autoencoders Distribution networks Forestry Nearest neighbor methods Robustness Leak detection Long short term memory Water resources

来源：评论

学校读者我要写书评

暂无评论

Experimental Evaluation of a Checklist-Based Inspection Technique to Verify the Compliance of Software systems with the Brazilian General Data Protection Law

arXiv

引用

arXiv 2023年

作者： Cerqueira, Diego André de Mello, Rafael Maiani Travassos, Guilherme Horta Systems Engineering and Computer Science Graduate Program Federal University of Rio de Janeiro Brazil Computing Institute Federal University of Rio de Janeiro Brazil

Background/Context: Recent laws to ensure the security and protection of personal data establish new software requirements. Consequently, new technologies are needed to guarantee software quality under the perception of privacy and protection of personal data. Therefore, we created a checklist-based inspection technique (LGPDCheck) to support the identification of defects in software artifacts based on the principles established by the Brazilian General Data Protection Law (LGPD). Objective/Aim: To evaluate the effectiveness and efficiency of LGPDCheck for verifying privacy and data protection (PDP) in software artifacts compared to ad-hoc techniques. Method: To assess LGPDCheck and ad-hoc techniques experimentally through a quasi-experiment (two factors, five treatments). The data will be collected from IoT-based health software systems built by software engineering students from the Federal University of Rio de Janeiro. The data analyses will compare results from ad-hoc and LGPDCheck inspections, the participant's effectiveness and efficiency in each trial, defects' variance and standard deviation, and time spent with the reviews. The data will be screened for outliers, and normality and homoscedasticity will be verified using the Shapiro-Wilk and Levene tests. Nonparametric or parametric tests, such as the Wilcoxon or Student's t-tests, will be applied as appropriate. © 2023, CC BY-NC-ND.

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

UAV-Based Optimization for Fruit Counting in Greenhouses

UAV-Based Optimization for Fruit Counting in Greenhouses

引用

International Conference on Unmanned Aircraft systems (ICUAS)

作者： João Vítor R. Vasconcelos Thayron M. Hudson André G. dos Santos Martin Saska Alexandre S. Brandão Department of Electrical Engineering Núcleo de Especialização em Robótica Graduate Program on Computer Science Federal Univer-sity of Vicosa Vicosa MG Brazil Graduate Program on Computer Science Federal University of Vicosa Vicosa MG Brazil Department of Cybernetics Multi-robot Systems Group Faculty of Electrical Engineering Czech Technical University in Prague Czech Republic

ISBN: (数字)9798350357882

ISBN: (纸本)9798350357899

This paper presents a solution for counting fruit in agricultural greenhouses using Unmanned Aerial Vehicles (UAV s). Initially, a heuristic based on Simulated Annealing was used to optimize the UAV's trajectory, ensuring efficient coverage of the beds. Next, digital image processing (DIP) techniques were implemented to count the fruit, including depth segmentation, application of bounding boxes, color filtering, and element counting. The DIP accuracy was evaluated in multiple scenarios and the results indicate high reliability in fruit counting, with the potential to optimize agricultural operations and provide valuable information to producers. Possible future improvements could include further refinements in image processing to increase the accuracy of counting other fruits. Ultimately, this work contributes to the advancement of automation in agriculture by offering a viable and efficient solution for counting fruit in greenhouses using UAV s.

关键词： Image segmentation Accuracy Image color analysis Greenhouses Simulated annealing Autonomous aerial vehicles Trajectory

来源：评论

学校读者我要写书评

暂无评论

Implementation of AI-Based Support Vector Machine Algorithm for Fraud Detection in Digital Payment systems

Implementation of AI-Based Support Vector Machine Algorithm ...

引用

International Seminar on Research of Information Technology and Intelligent systems (ISRITI)

作者： Edy Prayitno Rikie Kartadie Uning Lestari Munengsih Sari Bunga Accounting Information Systems Faculty of Information Technology Universitas Teknologi Digital Indonesia Yogyakarta Indonesia Computer Engineering Study Program Faculty of Information Technology Universitas Teknologi Digital Indonesia Yogyakarta Indonesia Informatics Study Program Faculty of Science and Information Technology Universitas Akprind Indonesia Yogyakarta Indonesia Department of Computer Engineering Smart City Information Systems Politeknik Negeri Indramayu Indramayu Indonesia

ISBN: (数字)9798331519643

ISBN: (纸本)9798331519650

The increasing use of digital payment systems has led to a rise in fraudulent activities, presenting a significant challenge in ensuring secure transactions. This research focuses on implementing the Support Vector Machine (SVM) algorithm with a Radial Basis Function (RBF) kernel to detect fraud in digital payment systems. One of the main challenges addressed in this study is the severe class imbalance in the dataset, where fraudulent transactions account for only 0.17% of total transactions. To overcome this, the SMOTE (Synthetic Minority Over-sampling Technique) method was applied to balance the dataset, allowing the model to better recognize fraudulent patterns. The results indicate that the SVM model achieved an accuracy of 99.93%, with a precision of 86.23% and a recall of 75.51%. These results demonstrate that SVM, combined with SMOTE and RBF kernel, is highly effective in detecting fraudulent transactions while minimizing false positives. This research provides a strong foundation for improving fraud detection models in the context of digital payment systems, offering enhanced security and trust for users. Further research could explore hybrid models and real-time data analysis to improve performance.

关键词： Support vector machines Seminars Adaptation models Accuracy Data models Real-time systems Fraud Security Kernel Intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Magnetoelastic coupling driven magnon gap in a honeycomb antiferromagnet

引用

Physical Review B 2024年第10期110卷 104419页

作者： Jeongheon Choe Andrey Baydin Gaihua Ye Bowen Ma Cynthia Nnokwe Martin Rodriguez-Vega David Lujan Swati Chaudhary Fuyang Tay Jiaming He Hiroyuki Nojiri Jianshi Zhou Gregory A. Fiete Junichiro Kono Rui He Xiaoqin Li Department of Physics Center of Complex Quantum Systems Center for Dynamics and Control of Materials Department of Electrical and Computer Engineering Smalley–Curl Institute Department of Physics Applied Physics Graduate Program Smalley–Curl Institute Department of Mechanical Engineering Institute for Materials Research Department of Physics and Astronomy Department of Materials Science and NanoEngineering

Cobalt titanate, CoTiO3, is a honeycomb antiferromagnet recently confirmed experimentally to host Dirac magnons, topological spin-orbit excitons, and chiral phonons. Here, we investigate a magnon gap at the zone center which calls for a refined spin Hamiltonian. We propose a microscopic model for the magnon gap and attribute it to a lattice-distortion (phonon)-induced higher-order spin interaction. Strong magnetoelastic coupling in CoTiO3 is also evident in Raman spectra, in which the magnetic order exerts a stronger influence on phonons corresponding to in-plane ionic motions than those with out-of-plane motions. We further examine the evolution of the zone-center magnons in a high magnetic field up to 18.5 T via THz absorption spectroscopy measurements. Based on this field dependence, we propose a spin Hamiltonian that not only agrees with magnon dispersion measured by inelastic neutron scattering but also includes fewer exchange constants and a realistic anisotropy term. Our work highlights the broad implications of magnetoelastic coupling in the study of topologically protected bosonic excitations.

关键词： Magnetism Magnons Topological materials Raman spectroscopy Terahertz spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Double-SP Method to Enhance Privacy and Security of Smart City Applications 10

Double-SP Method to Enhance Privacy and Security of Smart Ci...

引用

17th INDIACom;2023 10th International Conference on Computing for Sustainable Global Development, INDIACom 2023

作者： Bahbouh, Nour Mahmoud Hakami, Abdulelah Alhussain Hassan Aljohani, Yaser Mohammed Sen, Adnan Ahmed Abi Alkhodre, Ahmad Alsughayyir, Aeshah Granada University Department of Information and Communication Sciences Granada Spain Saudi Electronic University Al-Madinah Saudi Arabia YasRef Program Yanbu Saudi Arabia Computing and Information Technology Islamic University Al-Madinah Saudi Arabia Islamic University Faculty of Computer and Information Systems Al-Madinah Saudi Arabia College of Computer Science and Engineering Taibah University Al-Madinah Saudi Arabia

ISBN: (纸本)9789380544472

Smart cities promise a lot of well-being to their users in all areas of life through millions of applications and services. Smart services rely heavily on collecting data and the preferences of users. But on the dark side, the users' information is posed to threat and penetration during transmission or stored far in the clouds. Depending on encryption only is not sufficient if the attacker has strong resources or if the attacker is the service provider (SP) itself. In addition, changing data before sending is not a practical solution in many systems because of the adverse impact on the quality of a main service. This research presented a new idea to address the issue of protection. The proposed method enhanced the privacy and security of users' data without affecting the accuracy of service. The core of the suggested solution relies on a knowledge base of services that are managed by experts. Also, the solution depends on fog nodes to measure the level of security and privacy of users' queries without delay. Moreover, the fog nodes manage contact with SPs. Finally, the proposed method divided the SP into two, one for user queries and the other for user data. The simulation and analytical discussion on a practical case in smart cities demonstrated the superiority of the proposed approach over previous methods in the level of protection by maintaining the quality of services, and the resistance to attacks. © 2023 Bharati Vidyapeeth, New Delhi.

关键词： Fog

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：