检索结果-内蒙古大学图书馆

3rd Novel Intelligent and Leading Emerging Sciences Conference, NILES 2021

作者： Shaban, Nourhan M. Yousef, Ahmed H. Mohamed, Hoda K. Ain Shams University Department of Computer and Systems Engineering Cairo Egypt Egypt School of Information Technology and Computer Science Nile University Giza Egypt

ISBN: (纸本)9781665421577

Deployment on different platforms has been a great issue for mobile companies that aim to maximize the return on investments by making their mobile applications available on different mobile platforms. Consequently, the app may be developed several times to match different platforms. Therefore, there is a need to have solutions that enable the developers to develop the app once, and run it everywhere to reduce the cost of development and reach out to maximum users across several platforms. In this paper a tool is provided with the most popular languages /frameworks (native, cross platform) based on their market share to select the desired framework among them to code with, get the developed application written in the target language that meets their business requirements, and give the ability to convert native application to cross platform one. A trans-compiler based solution is introduced to convert the UI and UI-related backend code of application from a set of languages to any other language in the set and ability of adding more languages and frameworks in the future to achieve the concept of generalization of code conversion. © 2021 IEEE.

关键词： Android (operating system)

来源：评论

学校读者我要写书评

暂无评论

Employing Sentence Space Embedding for Classification of Data Stream from Fake News Domain

arXiv

引用

arXiv 2024年

作者： Zyblewski, Pawel Klikowski, Jakub Borek-Marciniec, Weronika Ksieniewicz, Pawel Department of Systems and Computer Networks Faculty of Information and Communication Technology Wroclaw University of Science and Technology Wybrzeże Wyspiańskiego 27 Wroclaw50-370 Poland

Tabular data is considered the last unconquered castle of deep learning, yet the task of data stream classification is stated to be an equally important and demanding research area. Due to the temporal constraints, it is assumed that deep learning methods are not the optimal solution for application in this field. However, excluding the entire – and prevalent – group of methods seems rather rash given the progress that has been made in recent years in its development. For this reason, the following paper is the first to present an approach to natural language data stream classification using the sentence space method, which allows for encoding text into the form of a discrete digital signal. This allows the use of convolutional deep networks dedicated to image classification to solve the task of recognizing fake news based on text data. Based on the real-life Fakeddit dataset, the proposed approach was compared with state-of-the-art algorithms for data stream classification based on generalization ability and time complexity. © 2024, CC BY-NC-SA.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An Application of Co-plot Analysis: A Multidimensional Scaling Data Visualization 23

An Application of Co-plot Analysis: A Multidimensional Scali...

引用

Proceedings of the 2023 6th International Conference on Machine Learning and Natural Language Processing

作者： Priyanka Poosapati Yao Shi Judith Gebauer Yang Song Department of Computer Science University of North Carolina Wilmington USA Congdon School of Supply Chain Business Analytics Information Systems University of North Carolina Wilmington USA

ISBN: (纸本)9798400709241

Exploratory graphs are a crucial element of statistical analysis and models that help us identify and interpret data patterns. As one example of exploratory graphs, multidimensional scaling (MDS) allows the visualization of large and complex datasets in a reduced dimensional space, while preserving pairwise differences and similarities between data points. Co-plot analysis extends the capabilities of traditional MDS and enables a multifaceted view of relationships between variables from different datasets. To develop a co-plot graph, the analyst starts out by standardizing the dataset and by generating a special coefficient (MADCC – median absolute deviation correlation coefficient) that subjects the dataset to MDS-embedding and helps to determine how the dataset fits the surface. In the current paper, we demonstrate how co-plot analysis can be applied to describe and compare the behavioral patterns of ecological communities and species across multiple sites and environmental contexts. We demonstrate the use of non-metric multidimensional scaling (NMDS) and robust multidimensional scaling (RMDS) techniques to visualize dissimilarities and distances of the observations in a reduced dimensional space. We also explain the use of Shepard graphs as a second technique to visualize categorical variables in a reduced-dimensional space, with points closer together indicating higher similarity. Shepard graphs also allow us to assess the goodness of fit, which is indicated by the strength of a positive linear relationship. For our application, the co-plots graphs are useful to visualize a complete set of co-relations between the species of different communities. It also allows us to plot how every factor affects the number of species against the mean and against the median and to determine the best approach: A co-plot graph against the mean may be influenced by the shape of the distribution, especially if the data is skewed, whereas a co-plot graph against the median is typica

关键词： Co-plot Data visualization Multidimensional scaling (MDS)

来源：评论

学校读者我要写书评

暂无评论

DeepVec: State-Vector Aware Test Case Selection for Enhancing Recurrent Neural Network

引用

IEEE Transactions on Software Engineering 2025年

作者： Jiang, Zhonghao Yan, Meng Huang, Li Sun, Weifeng Liu, Chao Sun, Song Lo, David Chongqing University School of Big Data and Software Engineering China Chongqing Normal University School of Computer and Information Science China Singapore Management University School of Information Systems Singapore

Deep Neural Networks (DNN) have realized significant achievements across various application domains. There is no doubt that testing and enhancing a pre-trained DNN that has been deployed in an application scenario is crucial, because it can reduce the failures of the DNN. DNN-driven software testing and enhancement require large amounts of labeled data. The high cost and inefficiency caused by the large volume of data of manual labeling, and the time consumption of testing all cases in real scenarios are unacceptable. Therefore, test case selection technologies are proposed to reduce the time cost by selecting and only labeling representative test cases without compromising testing performance. Test case selection based on neuron coverage (NC) or uncertainty metrics has achieved significant success in Convolutional Neural Networks (CNN) testing. However, it is challenging to transfer these methods to Recurrent Neural Networks (RNN), which excel at text tasks, due to the mismatch in model output formats and the reliance on image-specific characteristics. What’s more, balancing the execution cost and performance of the algorithm is also indispensable. In this paper, we propose a state-vector aware test case selection method for RNN models, namely DeepVec, which reduces the cost of data labeling and saves computing resources and balances the execution cost and performance. DeepVec selects data using uncertainty metric based on the norm of the output vector at each time step (i.e., state-vector), and similarity metric based on the direction angle of the state-vector. Because test cases with smaller state-vector norms often possess greater information entropy and similar changes of state-vector direction angle indicate similar RNN internal states. These metrics can be calculated with just a single inference, which gives it strong bug detection and model improvement capabilities. We evaluate DeepVec on five popular datasets, containing images and texts as well as commonl

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

Analyzing the behaviors of pedestrians and cyclists in interactions with autonomous systems using controlled experiments: A literature review

引用

Transportation Research Part F: Traffic Psychology and Behaviour 2025年 114卷 270-307页

作者： Danya Li Wencan Mao Francisco C. Pereira Yu Xiao Xiang Su Rico Krueger Department of Technology Management and Economics Technical University of Denmark (DTU) Denmark Information Systems Architecture Science Research Division National Institute of Informatics (NII) Japan Department of Information and Communications Engineering Aalto University Finland Department of Computer Science and Department of Agricultural Sciences University of Helsinki Finland

Urban transportation is set to undergo a profound transformation with the advent of autonomous systems such as autonomous vehicles, automated buses, and sidewalk delivery robots. To promote safe, sustainable, and inclusive urban mobility, understanding and predicting the behaviors of pedestrians and cyclists, including their intentions, decisions, and movements, when they interact with autonomous systems becomes crucial. Gaining a thorough understanding of these complex interactions can not only improve the safety, efficiency, and acceptance of autonomous systems but also enhance the design and implementation of these technologies. Through a comprehensive review of the literature spanning the years 2014 to 2023, we identify 99 articles that empirically investigate the interactions of humans and autonomous systems. Based on our overview of progress and challenges within this field, we further identify five research gaps that future research should address to enhance human-autonomous system interactions, including: (1) scaling up experimental scenarios to multi-user and multi-modal setups to better represent real-world challenges, (2) emphasizing safety-critical scenarios that are difficult to achieve in real-world environments by applying virtual reality, (3) incorporating diverse behavioral data from the human perspective to deepen the understanding of vulnerable road user behavior and decisions, (4) embracing continuous and real-time interaction to better predict dynamic future environments, and (5) enhancing the generalization ability to ensure realism and broad applicability. This review article offers valuable insights for the growing human-autonomous system research community, specifically those interested in leveraging controlled experiments to enhance the understanding and prediction of pedestrians' and cyclists' behaviors in future urban environments.

关键词： Human-autonomous system interaction Human behavioral modeling Autonomous vehicles Virtual reality Controlled experiment

来源：评论

学校读者我要写书评

暂无评论

Enhancing Wireless Sensor Network Data Collection through Aerial Unmanned Vehicles

Enhancing Wireless Sensor Network Data Collection through Ae...

引用

Decision Aid Sciences and Application (DASA), International Conference on

作者： Amjed Abbas Ahmed Khattab M. Ali Alheeti Ali Hamza Najim Nurhizam Safie Mohd Satar Salah A. Aliesawi Aisha Hassan Abdallah Hashim Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Bangi Malaysia Department of Computer Techniques Engineering mam AlKadhum College (IKC) Baghdad Iraq Computer Networking Systems Department College of Computer and Information Technology University of Anbar Anbar Iraq Department of Computer Technical Engineering Imam Al-Kadhum College (IKC) Al-Diwaniyah Iraq Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia (UKM) Bangi Malaysia Department of Computer Networking Systems College of Computer and Information Technology University of Anbar Anbar Iraq Department of Electrical and Computer Engineering Faculty of Engineering International Islamic University Malaysia Selangor Malaysia

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

This study addresses the critical aspect of data collection within Wireless Sensor Networks (WSNs), which consist of autonomous, compact sensor devices deployed to monitor environmental conditions. These networks have limited processing capacity, storage, short-range communication capabilities, and constrained battery power. A primary challenge in WSN s revolves around data gathering, particularly when employing static sinks, leading to premature energy depletion in nodes near these sinks. Integrating mobile collectors introduces the potential for energy conservation by shortening data routes. This article introduces three distinct algorithms for optimizing data collection in WSN s. Our approach involves an in-depth analysis of diverse strategies to mitigate packet collisions and facilitate the retransmission of lost packets. Our findings underscore the considerable impact that packet collisions can exert on the overall network performance during data gathering within wireless sensor networks. Through empirical exploration and analysis, we highlight the significance of these algorithms in enhancing the efficiency, energy conservation, and reliability of data collection processes within WSNs. This research sheds light on these algorithms' vital role in addressing the pressing challenges of data gathering and network performance optimization in wireless sensor networks.

关键词： Wireless communication Wireless sensor networks Protocols Energy conservation Pressing Data collection Reliability Optimization Monitoring Drones

来源：评论

学校读者我要写书评

暂无评论

Lockstep Vs Microarchitecture: A Comparison

Lockstep Vs Microarchitecture: A Comparison

引用

IEEE International SOC Conference

作者： Ján Mach Lukáš Kohútka Pavel Čičák Institute of Computer Engineering and Applied Informatics Slovak University of Technology Bratislava Slovakia Institute of Informatics Information Systems and Software Engineering Slovak University of Technology Bratislava Slovakia

ISBN: (数字)9798350377569

ISBN: (纸本)9798350377576

Safety/mission-critical applications require high dependability of the control systems. Their state-of-the-art protection approach is a system-level lockstep. This paper compares the system-level dual and triple lockstep technique to the microarchitecture-level protection of the Hardisc RISC-V core. For a fair comparison, each system is based on the same core(s) and integrates protection against bit-flips in memory and transient faults in the bus. We propose a fault injection methodology, combining pre-synthesis simulation with synthesis data to analyse the vulnerability of a system to faults. The fault injection campaigns show that the Hardisc can withstand fault rates orders of magnitude higher than the dual-core lockstep system while preserving the same area and power consumption. It comes with a 5% frequency penalty. We have shown that system failures are more frequent once the fault rate reaches an application-specific threshold (one fault in 100 clock cycles).

关键词： Fault tolerance Microarchitecture Power demand Event detection Fault tolerant systems Registers System-on-chip Protection Transient analysis Clocks

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Convolutional Neural Networks Classification Based on Brain Cancer MRI

Deep Learning Convolutional Neural Networks Classification B...

引用

2023 International Conference on Electrical, computer, Communications and Mechatronics Engineering, ICECCME 2023

作者： Zeebaree, Diyar Qader Asaad Zebari, Dilovan Zeebaree, Imad Majed Abduljabbar Mohammed, Mohammed Mohammed, Bafreen Abduljabbar Asaad Zebari, Nechirvan Duhok Polytechnic University Technical College of Informatic Department of Computer Network and Information Security Kurdistan Region Duhok42001 Iraq Nawroz University College of Science Department of Computer Science Kurdistan Region Duhok42001 Iraq Duhok Private Technical Institute Department of Health Management Information Systems Kurdistan Region Duhok42001 Iraq University of Duhok College of Education Department of Mathematic Kurdistan Region Duhok42001 Iraq

ISBN: (纸本)9798350322972

Brain tumor diagnosis and classification is a major challenge for medical field. The majority of the time, computer-Aided Diagnostic systems (CADS) and methods involving Deep learning (DL) are utilized to find brain tumors quickly and precisely. Artificial diagnostic systems, on the other hand, are not very accurate, the physician is not using them very well when diagnosing Magnetic Resonance Images (MRI). By applying image processing principles, it is possible to conceptualize the different human tissue structures. It is difficult to detect diseased structures in the human brain using only the most fundamental imaging methods. In this research, we present a model-based Convolutional Neural Network (CNN) structure for identifying and classifying brain tumors. The proposed method consists of two parts: pre-processing and classification. In this study, CNN architecture outperformed in state-of-the-art methods, with 99.87% accuracy. F1-score 99.88%, 99.76% precision. © 2023 IEEE.

关键词： Magnetic resonance

来源：评论

学校读者我要写书评

暂无评论

Scenario-Based Field Testing of Drone Missions

Scenario-Based Field Testing of Drone Missions

引用

Euromicro Conference on Software Engineering and Advanced Applications (SEAA)

作者： Michael Vierhauser Kristof Meixner Stefan Biffl Department of Computer Science University of Innsbruck Austria CDL-SQI Vienna Austria Institute of Information Systems Engineering TU Wien Vienna Austria CDP Vienna Austria

ISBN: (数字)9798350380262

ISBN: (纸本)9798350380279

Testing and validating Cyber-Physical systems (CPSs) in the aerospace domain, such as field testing of drone rescue missions, poses challenges due to volatile mission environments, such as weather conditions. While testing processes and methodologies are well established, structured guidance and execution support for field tests are still weak. This paper identifies requirements for field testing of drone missions and introduces the Field Testing Scenario Management (FiTS) approach for adaptive field testing guidance. FiTS aims to provide sufficient guidance for field testers as a foundation for efficient data collection to facilitate quality assurance and iterative improvement of field tests and CPSs. FiTS shall leverage concepts from scenario-based requirements engineering and Behavior-Driven Development to define structured and reusable test scenarios, with dedicated tasks and responsibilities for role-specific guidance. We evaluate FiTS by (i) applying it to three use cases for a search-and-rescue drone application to demonstrate feasibility and (ii) interviews with three experienced drone developers to assess its usefulness and collect further requirements. The study results indicate FiTS to be feasible and useful to facilitate drone field testing and data analysis.

关键词： Quality assurance Navigation Data collection Requirements engineering Iterative methods Interviews Testing Drones Software engineering Meteorology

来源：评论

学校读者我要写书评

暂无评论

Enhancing 3D Human Pose Estimation Amidst Severe Occlusion with Dual Transformer Fusion

arXiv

引用

arXiv 2024年

作者： Ghafoor, Mehwish Mahmood, Arif Bilal, Muhammad Department of Computer Science Information Technology University Lahore Pakistan Center of Excellence in Intelligent Engineering Systems Department of Electrical and Computer Engineering Faculty of Engineering King Abdulaziz University Jeddah21589 Saudi Arabia

In the field of 3D Human Pose Estimation from monocular videos, the presence of diverse occlusion types presents a formidable challenge. Prior research has made progress by harnessing spatial and temporal cues to infer 3D poses from 2D joint observations. This paper introduces a Dual Transformer Fusion (DTF) algorithm, a novel approach to obtain a holistic 3D pose estimation, even in the presence of severe occlusions. Confronting the issue of occlusion-induced missing joint data, we propose a temporal interpolation-based occlusion guidance mechanism. To enable precise 3D Human Pose Estimation, our approach leverages the innovative DTF architecture, which first generates a pair of intermediate views. Each intermediate-view undergoes spatial refinement through a self-refinement schema. Subsequently, these intermediate-views are fused to yield the final 3D human pose estimation. The entire system is end-to-end trainable. Through extensive experiments conducted on the Human3.6M and MPI-INF-3DHP datasets, our method’s performance is rigorously evaluated. Notably, our approach outperforms existing state-of-the-art methods on both datasets, yielding substantial improvements. The code is available here: https://***/MehwishG/DTF. © 2024, CC BY-NC-ND.

关键词： 3D modeling

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：