检索结果-内蒙古大学图书馆

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Chen, Ke Han, Xiaosong Li, Xiaoran Liang, Yanchun Xu, Dong Guan, Renchu Jilin University Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Software Changchun China Jilin University Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Computer Science and Technology Changchun China Zhuhai College of Science and Technology Zhuhai Laboratory of Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education School of Computer Science Zhuhai China University of Missouri Christopher S. Bond Life Sciences Center Department of Electrical Engineering and Computer Science Columbia United States

ISBN: (纸本)9798350386226

Drug-Drug Interaction (DDI) task plays a crucial role in clinical treatment and drug development. Recently, deep learning methods have been successfully applied for DDI prediction. However, training deep learning models always need large amount of data, while known DDIs are scarce. To address this challenge, a graph neural network-based DDI prediction model named PNESR-DDI is proposed, which compensates for the lack of DDIs by enriching drug representations. First, to obtain initial node representations that incorporate rich semantic information from the biomedical knowledge graph (KG), a link prediction pre-training method on external KG is proposed in the node embedding pre-training module. Then, considering the large scale of the KG, subgraph extraction for the target drug pairs is introduced to reduce noise and decrease computational complexity in the subgraph anchoring module. After that, the subgraph is updated, and node similarities are propagated in the subgraph reconstruction module. Based on the node similarity scores, the subgraph is pruned and reconstructed, which adjusts node representations to be more conducive to DDI prediction. Finally, the drug embeddings, subgraph representations, and drug fingerprint features are concatenated to predict DDIs. PNESRDDI is evaluated on two benchmark DDI datasets: DrugBank and TWOSIDES. Experiment results show that PNESR-DDI achieves better performance than baselines. Ablation results validate the effectiveness of the pre-training method and the adaptive subgraph reconstruction strategy. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Synthetic Aircraft Trajectory Generation Using Time-Based VQ-VAE

arXiv

引用

arXiv 2025年

作者： Murad, Abdulmajid Ruocco, Massimiliano Department of Software Engineering Safety and Security SINTEF Digital Trondheim Norway Department of Computer Science Norwegian University of Science and Technology Trondheim Norway

In modern air traffic management, generating synthetic flight trajectories has emerged as a promising solution for addressing data scarcity, protecting sensitive information, and supporting large-scale analyses. In this paper, we propose a novel method for trajectory synthesis by adapting the Time-Based Vector Quantized Variational Autoencoder (TimeVQVAE). Our approach leverages time-frequency domain processing, vector quantization, and transformer-based priors to capture both global and local dynamics in flight data. By discretizing the latent space and integrating transformer priors, the model learns long-range spatiotemporal dependencies and preserves coherence across entire flight paths. We evaluate the adapted TimeVQVAE using an extensive suite of quality, statistical, and distributional metrics, as well as a flyability assessment conducted in an open-source air traffic simulator. Results indicate that TimeVQVAE outperforms a temporal convolutional VAE baseline, generating synthetic trajectories that mirror real flight data in terms of spatial accuracy, temporal consistency, and statistical properties. Furthermore, the simulator-based assessment shows that most generated trajectories maintain operational feasibility, although occasional outliers underscore the potential need for additional domain-specific constraints. Overall, our findings underscore the importance of multi-scale representation learning for capturing complex flight behaviors and demonstrate the promise of TimeVQVAE in producing representative synthetic trajectories for downstream tasks such as model training, airspace design, and air traffic forecasting. © 2025, CC BY.

关键词： Aerobraking

来源：评论

学校读者我要写书评

暂无评论

Research on microseismic source localization based on improved particle swarm optimization algorithm

Research on microseismic source localization based on improv...

引用

Artificial Intelligence, Robotics, and Communication (ICAIRC), IEEE International Conference on

作者： Yunshuai Wan Jinchun Xue Hui Qi Feipeng Ren School of Energy and Mechanical Engineering Jiangxi University of Science and Technology Nanchang China School of Software Engineering Jiangxi University of Science and Technology Nanchang China School of Information Engineering Shandong Shenghan Finance and Trade Vocational College Jinan China

ISBN: (数字)9798331531225

ISBN: (纸本)9798331531232

Accurate microseismic source localization is crucial for detecting potential geological hazards during mining operations, ensuring the safe and efficient progress of excavation activities. The accuracy and reliability of source localization are closely linked to the selection of appropriate algorithms. While the Particle Swarm Optimization (PSO) algorithm offers the advantage of fast convergence, it is prone to premature convergence and low localization accuracy when directly applied to source localization. To address these issues, this study employs chaotic mapping to replace traditional methods for initializing the particle population, enhancing population diversity. The proposed method is validated through engineering case studies. The results demonstrate that, compared to the traditional PSO algorithm, the chaotic mapping-based improved PSO algorithm achieves faster convergence and superior optimization capabilities, making it suitable for solving practical engineering problems.

关键词： Location awareness Accuracy Chaotic mapping Reliability Particle swarm optimization Standards Optimization Robots Convergence Logistics

来源：评论

学校读者我要写书评

暂无评论

A Highly Accurate Data Synchronization and Full-text Search Algorithm for Canal and Elasticsearch 20

A Highly Accurate Data Synchronization and Full-text Search ...

引用

20th IEEE International Conference on Networking, Sensing and Control, ICNSC 2023

作者： Wei, Peiyang Shi, Xiaoyu Zhang, Gang Chongqing University of Posts and Telecommunications School of Software Engineering Chengdu University of Information Technology School of Computer Science and Technology Chongqing China University of Chinese Academy of Sciences Chongqing Institute of Green and Intelligent Technology Chinese Academy of Sciences Chongqing School Chongqing China Cnnc Center for Fusion Science Southwestern Institute of Physics Chengdu China

ISBN: (纸本)9798350369502

Currently, there are numerous thorny issues in structured data and semi-structured full-text search scheme with large-scale text nature, like long data synchronization delay, inconvenient personalized business processing and low efficiency. To address these issues, this paper proposes an efficient algorithm based on Canal data synchronization framework and Elasticsearch full-text search engine. Firstly, we rewrite the Canal adapter component to obtain the flexible configuration of business data processing, thereby enhancing the secondary data processing ability of the framework and achieving the purpose of improving the efficiency of data synchronization. Secondly, by recording the synchronization time of nearby data in Canal framework, the weight of time series data is gradually decreased by combining with the exponential weighted average function to highlight the influence of recent data and present the novelty of data, which can achieve effective control the synchronization interval and duration by dynamically and flexibly setting the synchronization trigger period. Lastly, the Elasticsearch word tokenizer is modified, and then the configuration of custom expansion words and stop words dictionary are proposed to filter the query data effectively, thereby enhancing the query hit rate and accuracy. Extensive experiments on the data of traditional Chinese medicine demonstrate that the designed algorithm obtains high data synchronization efficiency, full text search speed and accuracy. Hence, the proposed algorithm is a milestone in smart healthcare. © 2023 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

SAFEJ: An Efficient Infrastructure for Securing Julia Programs

SAFEJ: An Efficient Infrastructure for Securing Julia Progra...

引用

Big Data & Artificial Intelligence & software engineering (ICBASE), International Conference on

作者： Hao Zhu Baojian Hua School of Software Engineering Suzhou Institute for Advanced Research University of Science and Technology of China Suzhou China

Julia is an emerging programming language designed for high-performance and parallel computing. However, its simple-designed foreign function interface, ccall, lacks necessary security checks, making multilingual Julia applications vulnerable to exploitation. This paper introduces SAFEJ, an infrastructure developed to enhance Julia FFI security. SAFEJ includes two key components: 1) privilege separation to protect Julia by isolating the memory of Julia and native code and 2) pointer sanitization to filter and check memory accesses from unsafe native code to provide more flexible protection. We conducted extensive experiments to evaluate the effectiveness and performance of SAFEJ. The results demonstrate that SAFEJ effectively protects Julia from common memory attacks, and its efficiency tests indicate that pointer sanitization is on par with prior studies, while privilege separation achieves a maximum reduction of 85% in runtime performance overhead.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Person Re-Identification with Model-Contrastive Federated Learning in Edge-Cloud Environment

引用

Intelligent Automation & Soft Computing 2023年第10期38卷 35-55页

作者： Baixuan Tang Xiaolong Xu Fei Dai Song Wang School of Computer and Software Nanjing University of Information Science and TechnologyNanjing210044China Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET) Nanjing University of Information Science and TechnologyNanjing210044China School of Big Data and Intelligence Engineering Southwest Forestry UniversityKunming650233China Nanjing Meteorological Service Center Nanjing Meteorological BureauNanjing210019China

Person re-identification(ReID)aims to recognize the same person in multiple images from different camera *** person ReID models are time-consuming and resource-intensive;thus,cloud computing is an appropriate model training ***,the required massive personal data for training contain private information with a significant risk of data leakage in cloud environments,leading to significant communication *** paper proposes a federated person ReID method with model-contrastive learning(MOON)in an edge-cloud environment,named ***,based on federated partial averaging,MOON warmup is added to correct the local training of individual edge servers and improve the model’s effectiveness by calculating and back-propagating a model-contrastive loss,which represents the similarity between local and global *** addition,we propose a lightweight person ReID network,named multi-branch combined depth space network(MB-CDNet),to reduce the computing resource usage of the edge device when training and testing the person ReID ***-CDNet is a multi-branch version of combined depth space network(CDNet).We add a part branch and a global branch on the basis of CDNet and introduce an attention pyramid to improve the performance of the *** experimental results on open-access person ReID datasets demonstrate that FRM achieves better performance than existing baseline.

关键词： Person re-identification federated learning contrastive learning

来源：评论

学校读者我要写书评

暂无评论

AdvCLIP: Downstream-agnostic Adversarial Examples in Multimodal Contrastive Learning

arXiv

引用

arXiv 2023年

作者： Zhou, Ziqi Zhang, Hangtao Hu, Shengshan Zhang, Yechao Li, Minghui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China

Multimodal contrastive learning aims to train a general-purpose feature extractor, such as CLIP, on vast amounts of raw, unlabeled paired image-text data. This can greatly benefit various complex downstream tasks, including cross-modal image-text retrieval and image classification. Despite its promising prospect, the security issue of cross-modal pre-trained encoder has not been fully explored yet, especially when the pre-trained encoder is publicly available for commercial use. In this work, we propose AdvCLIP, the first attack framework for generating downstream-agnostic adversarial examples based on cross-modal pre-trained encoders. AdvCLIP aims to construct a universal adversarial patch for a set of natural images that can fool all the downstream tasks inheriting the victim cross-modal pre-trained encoder. To address the challenges of heterogeneity between different modalities and unknown downstream tasks, we first build a topological graph structure to capture the relevant positions between target samples and their neighbors. Then, we design a topology-deviation based generative adversarial network to generate a universal adversarial patch. By adding the patch to images, we minimize their embeddings similarity to different modality and perturb the sample distribution in the feature space, achieving unviersal non-targeted attacks. Our results demonstrate the excellent attack performance of AdvCLIP on two types of downstream tasks across eight datasets. We also tailor three popular defenses to mitigate AdvCLIP, highlighting the need for new defense mechanisms to defend cross-modal pre-trained encoders. Our codes are available at: https://***/CGCL-codes/AdvCLIP. Copyright © 2023, The Authors. All rights reserved.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Leveraging the Internet of Things and Cloud Computing: A Crowdsourcing Web Application for Reporting Hazard Incidents

Leveraging the Internet of Things and Cloud Computing: A Cro...

引用

AFRICON Conference

作者： Reginald E. Ogu Gloria A. Chukwudebe Alex Watila Uche M. Mbanaso Chinomso D. Okoronkwo Nkwachukwu Chukwuchekwa Software Engineering Federal University of Technology Owerri Owerri Nigeria Electrical and Electronic Engineering Federal University of Technology Owerri Owerri Nigeria KASNEB Nairobi Kenya Center for Cyberspace Studies Nasarawa State University Keffi Keffi Nigeria Software Engineering Dept. Federal University of Technology Owerri Owerri Nigeria

Crowdsourcing Web Application has not been utilized effectively for hazard identification. This paper presents the development of a Web-based crowdsourcing platform for hazard identification. Hence, a Web Application based on the Model- View-Controller framework is presented in this paper. The Web Application provides a platform through which data from IoT devices monitoring the environment for hazardous air pollutants can be visualized. Also, human users are enabled to report other forms of hazard around them through the crowdsourcing capability of the Web Application. The Laravel PHP framework is employed for the backend part of the Application while a combination of HyperText Markup Language, Tailwind Cascading Style Sheets and J avaScript were employed in the frontend part. JavaScript Object Notation is used as the data exchange format between the Internet of Things devices, cloud database and Web Application. Upon testing, the Application was confirmed to be working in line with the requirements gathered at the beginning of the development process. Notably, the IoT sensor data was visualizable through the Web Application in a user-friendly manner and users were able to communicate hazardous incidents using the App. Thus, the Web Application facilitates the hazard identification task, thereby improving the risk management processes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Incremental Instance Segmentation for Cluttered Baggage Threat Detection

Incremental Instance Segmentation for Cluttered Baggage Thre...

引用

2023 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications, CIVEMSA 2023

作者： Nasim, Ammara Velayudhan, Divya Ahmed, Abdelfatah Hassan Hassan, Taimur Akcay, Samet Akram, Muhammad Usman Werghi, Naoufel National University of Sciences and Technology Department of Computer and Software Engineering Pakistan Khalifa University KUCARS and C2PS Department of Electrical Engineering and Computer Science United Arab Emirates Abu Dhabi University Department of Electrical and Computer Engineering Abu Dhabi United Arab Emirates Intel Corporation RandD Division United Kingdom

ISBN: (纸本)9781665434454

Identification of contraband items from highly oc-cluded baggage of air travelers is a challenging task even for human experts with very high experience. Many researchers have been working rigorously to develop computer vision-based techniques for baggage screening through X-ray images. Nu-merous machine learning and deep learning-based frameworks have been proposed by researchers in the last two decades. However, all of these techniques face limitations in segmenting prohibited items from highly occluded and cluttered baggage. In this paper, we propose a novel framework based on semantic segmentation to automatically detect concealed prohibited items from X-ray baggage scans. Furthermore, to detect different overlapping instances of the same contraband item, we propose an instance-Aware segmentation model that enables the semantic segmentation model to identify multiple instances of the same threat category through incremental learning without requiring additional overhead. The proposed framework is computationally lighter compared to other similar approaches as it requires min-imal training examples and leverages previous knowledge. The proposed model has outperformed state-of-The-Art instance seg-mentation techniques when tested on publicly available GDXray and SIXray datasets, giving mean average precision scores of 0.50 and 0.47, respectively. In addition, the proposed framework leads other instance segmentation baseline models in terms of mean inference time. © 2023 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

AI-Powered Conceptual Model for Scrum Framework

AI-Powered Conceptual Model for Scrum Framework

引用

Computing and Information technology (ICCIT-1441), 2020 International Conference on

作者： Rana Mdallal Dania Alsawalmeh Dania Jarad Ammar Odeh Department of Software Engineering/Enterprise Systems Engineering (ESE) Princess Sumaya University for Technology (PSUT) Department of Computer Science/Cybersecurity Princess Sumaya University for Technology (PSUT)

Scrum is an incremental, interactive agile framework widely used to develop practical problems; the Scrum team focuses on achieving a common goal each cycle, "Sprint Goal" instead of the traditional sequential technique for delivering products. Although many organizations are adopting Scrum, it is still following the traditional manual approaches. Even though many tools are available in the market to facilitate project management and Scrum master tasks, the support is still limited and lacks the utilization of Artificial Intelligence science. This study focuses on integrating the Scrum framework with Artificial Intelligence to build a predicting model that can assist project management "in particular the Scrum master" to predict the achievement of the Scrum Team per Sprint using Story Points; this prediction is built based on several factors that can affect the burndown charts per Sprint.

关键词：

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：