检索结果-内蒙古大学图书馆

4th international conference on Computational Methods in Science and Technology, ICCMST 2024

作者： Kabilan, K. Jebathangam, J. Department of Computer Science VISTAS Chennai India

ISBN: (纸本)9781032911571

Air pollution is a major health and environmental issue in increasingly urbanizing countries like India. This study examines ML and DL implementations on Indian Air Quality Index (AQI) datasets from 2015 to 2020. The research seeks to identify patterns, anticipate trends, and guide air pollution mitigation decisions. The study carefully analyses AQI parameters, explores data, and uses powerful ML and DL algorithms to forecast AQI and identify pollution hotspots. Technical features include missing value imputation, feature engineering, model selection criteria, hyper parameter tuning procedures, and model performance assessment measures. Key results show that ML and DL models can accurately forecast air quality, identify pollution sources, and capture temporal and geographical fluctuations. The article suggests integrating more data sources, exploring advanced deep learning methods, and developing dynamic air quality control systems. This study uses data-driven insights and multidisciplinary techniques to improve our knowledge of air quality dynamics in India and create the framework for proactive air pollution mitigation and public health policies. © 2025 the Author(s).

关键词： data quality

来源：评论

学校读者我要写书评

暂无评论

HA-MARL: Heuristic and APF Assisted Multi-agent Reinforcement Learning for Wireless data Sharing in AUV Swarms 59

HA-MARL: Heuristic and APF Assisted Multi-agent Reinforcemen...

引用

59th Annual IEEE international conference on Communications (IEEE ICC)

作者： Li, Zonglin Du, Jun Jiang, Chunxiao Mi, Weishi Ren, Yong Tsinghua Univ Tsinghua Shenzhen Int Grad Sch Shenzhen 518055 Peoples R China Tsinghua Univ Dept Elect Engn Beijing 100083 Peoples R China Tsinghua Univ Tsinghua Space Ctr Beijing 100083 Peoples R China

ISBN: (纸本)9781728190549

This paper focuses on the design of intelligent game strategy for multi-autonomous underwater vehicle (multi-AUV) underwater network system. The challenge lies in ensuring the coordination and stability between AUVs in complex underwater environments. To meet underwater data sharing requirements, we formulate an intelligent game strategy incorporating communication delays by formulating the problem as a partially observable Markov decision process (POMDP). Additionally, to address the issue of sparse rewards during exploration in multi-agent reinforcement learning (MARL) models and improve the coordination among AUVs, we propose a heuristic and artificial potential field (APF)-assisted multi-agent proximal policy optimization (HA-MAPPO) algorithm. Our proposed scheme addresses the issue of sparse rewards in MARL by using APF as path planner and subsequently utilizes heuristic algorithm for task scheduling to achieve optimal goal allocation. Simulation results demonstrate that our proposed HA-MAPPO algorithm outperforms current mainstream MARL algorithms regarding convergence speed while maximizing the winning rates.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Influence of COVID-19 in Tourism: A Sentiment Analysis of Twitter data 4

Influence of COVID-19 in Tourism: A Sentiment Analysis of Tw...

引用

4th international conference on Sustainable Expert Systems, ICSES 2024

作者： Kumaravel, T. Nallusamy, C. Natesan, P. Nagul, K.S. Naveen, A.M. Sakthisundaram, A. Kongu Engineering College Department of Computer Science and Engineering Tamil Nadu Erode India K S Rangasamy College of Technology Dept of information Technology TamilNadu Namakkal India

ISBN: (纸本)9798331540364

This study investigates the economic impacts of the COVID-19 pandemic on Thailand's tourism-driven economy by analyzing English-language tweets related to tourism from July to December 2020. Utilizing advanced machine learning models-Logistic Regression, Decision Tree Classifier, XGBoost Classifier, and the Pelican Optimization Algorithm-we aimed to understand the sentiments and intentions expressed in these tweets. The research analysis revealed high predictive accuracy, with Logistic Regression achieving 98%, Decision Tree Classifier 99%, XGBoost Classifier 96%, and Pelican Optimization Algorithm 94.5%. These results highlight the importance of addressing concerns related to COVID-19 and political tensions to support the recovery of Thailand's tourism sector. The research also highlights the critical role of information and communications technology (ICT) in managing and revitalizing the tourism industry, especially in the context of reduced consumer spending and travel restrictions. Social media platforms, particularly Twitter, were used to gauge public sentiment and inform recovery strategies. The proposed methodology included extensive data collection, cleaning, and natural language processing, followed by the application of various machine learning algorithms to analyze and interpret sentiment data. The findings offer actionable insights for enhancing tourism marketing strategies and improving the sector's resilience amid ongoing pandemic challenges. © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Generating Event Logs from Hybrid Process models

Generating Event Logs from Hybrid Process Models

引用

21st international conference on Business Process Management (BPM)

作者： Alman, Anti Maggi, Fabrizio Maria Montali, Marco Rivkin, Andrey Univ Tartu Inst Comp Sci Tartu Estonia Free Univ Bozen Bolzano Bolzano Italy Tech Univ Denmark Inst Math & Comp Sci Lyngby Denmark

ISBN: (纸本)9783031509735;9783031509742

To carry on controlled experiments in process mining, it is necessary to generate event logs with specific characteristics. This has led to the development of log generation techniques in which a process model is simulated to generate an event log that is both compliant with the process model and also has certain user-defined properties (e.g., a certain number of traces, traces with certain lengths, etc.). Such techniques are available for a variety of modeling languages, both procedural and declarative. However, they are limited to simulating a single (procedural or declarative) process model at a time and do not allow simulating concurrent executions of multiple separate, but interacting, processes. In this paper, we introduce a log generation approach that takes multiple (procedural and declarative) process models (i.e., a Hybrid Business Process Representation) as input and produces an event log matching the concurrent execution of these models on the same case instances. We discuss the details of our approach and evaluate its implementation.

关键词： Log Generation Process Simulation Hybrid Process Model data Petri net Declare Finite State Automaton

来源：评论

学校读者我要写书评

暂无评论

MAID: Model Attribution via Inverse Diffusion

MAID: Model Attribution via Inverse Diffusion

引用

2025 IEEE international conference on Acoustics, Speech, and Signal processing, ICASSP 2025

作者： Zhu, Luyu Ye, Kai Yao, Jiayu Li, Chenxi Zhao, Luwen Cao, Yuxin Wang, Derui Hao, Jie Beijing University of Posts and Telecommunications China The University of Hong Kong Hong Kong National University of Singapore Singapore CSIRO's Data61 Australia

ISBN: (纸本)9798350368741

The surge in AI-generated images, blending authentic and synthetic content, raises security concerns and complicates model attribution, especially with limited transparency. Existing methods either struggle to attribute across multiple frameworks or rely on additional conditions, such as textual descriptions and white-box access to the source model, limiting their practicality and effectiveness in real-world scenarios. To address this gap, we introduce Model Attribution via Inverse Diffusion (MAID), the first framework-agnostic and self-sufficient approach that leverages the source model features extracted by diffusion models, which also works for images generated from GANs. By employing the inverse diffusion process, we are able to utilize pre-trained Diffusion models as Denoising Autoencoders, mapping images into a latent space and extracting the Diffusion Model Activations (DMA). This mapping effectively captures the unique characteristics of images originating from different source models, including authentic images, which showcase distinct latent Gaussian signatures. Experimental results show that, even in data-asymmetric unfair comparisons, the attribution classifier trained with our proposed DMA achieves approximately 15% and 3% higher ACC compared to SOTA methods on the DiffusionForensics and Artifact datasets, respectively. The code is available at https://***/Zhu-Luyu/MAID. © 2025 IEEE.

关键词： Deepfake Diffusion Model Model Attribution

来源：评论

学校读者我要写书评

暂无评论

Gap and Glitch Simulation for Taiji data Challenge 6

Gap and Glitch Simulation for Taiji Data Challenge

引用

6th international conference on Information Communication and Signal processing, ICICSP 2023

作者： Jiafeng, Zhang Xiaodong, Peng Mengyuan, Zhao Peng, Xu Pengzhan, Wu Chen, Gao Xiaoshan, Ma National Space Science Center Chinese Academy of Sciences Key Laboratory of Electronics and Information Technology for Space System Beijing China Institute of Mechanics Chinese Academy of Sciences Beijing China Institute of Engineering Thermophysics Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798350339994

Simulating the detection process of a space-based gravitational wave (GW) measurement system is crucial for the task demonstration, design, operation and data processing. Gaps and glitches, which are common types of abnormal events in the detection process, are an important part of data simulation. This paper explores two gap models and three glitch models, provides a comprehensive overview of these models. The implementation of these models in Taiji data, the simulation data of the Taiji program, is designed. Dedicated software modules were developed to incorporate the gap and glitch models into the simulation process. The impact of adding gap and glitch to the measurement is illustrated and analyzed. Finally, the time-series simulation detection data of space GWs that integrates gap and glitch models has been obtained, providing support for promoting the improvement of the detection data processing capabilities. © 2023 IEEE.

关键词： Gap simulation Glitch simulation Space gravitational wave detection Taiji data

来源：评论

学校读者我要写书评

暂无评论

Risk Analysis of Enterprise Cross-network data via DAE-BiLSTM and Fuzzy Bayesian Network 4

Risk Analysis of Enterprise Cross-network Data via DAE-BiLST...

引用

4th international conference on Intelligent Communications and Computing

作者： Li, Wei Li, Jianjun Shan, Yuxiang Li, Junjie Nie, Guibing China Tobacco Zhejiang Ind Co Ltd Informat Ctr Hangzhou Peoples R China Zhejiang Univ Sch Engn Hangzhou Peoples R China Hangzhou DBAPPSecur Co Ltd Hangzhou Peoples R China

ISBN: (纸本)9798331530372;9798331530365

With the increasing severity of enterprise information security issues, the identification and evaluation of cross-network data risk has become one of the essential technologies to safeguard enterprise data and network security. However, traditional rule-based or signature-based anomaly detection methods are increasingly unable to cope with the complex and evolving forms of attacks, necessitating more intelligent and flexible detection mechanisms to counter these constantly changing threats. To address this, a method integrating denoising autoencoder (DAE) and bidirectional long short-term memory network (BiLSTM) is first proposed for monitoring the enterprise cross-network data traffic and identify the occurring abnormal risks. On this basis, we use a fuzzy Bayesian network to effectively measure these known abnormal risks. In the end, experimental results demonstrate that the proposed method outperforms traditional machine learning algorithms and other deep learning models across various evaluation metrics, exhibiting high accuracy and efficiency.

关键词： network security cross-network data risk analysis deep learning

来源：评论

学校读者我要写书评

暂无评论

Key technologies and applications of intelligent design for complex traffic node driven by mathematics-modeling

引用

international JOURNAL OF MODERN PHYSICS C 2025年第6期36卷

作者： Yang, Yukui Zhang, Jiyong Chen, Weili Zhai, Cong Guangzhou Urban Planning & Design Survey Res Inst Guangzhou 510060 Peoples R China Guangdong Enterprise Key Lab Urban Sensing Monitor Guangzhou 510060 Peoples R China Collaborat Innovat Ctr Nat Resources Planning & Ma Guangzhou 510060 Peoples R China Guangzhou Rd Affairs Ctr Guangzhou 510060 Peoples R China Foshan Univ Sch Transportat & Civil Engn & Architecture Foshan 528000 Peoples R China

The intelligent design of complex transportation nodes is an important way to improve urban transportation efficiency and safety. However, currently, there are still problems with low data acquisition and processing accuracy and high-technology application costs in complex transportation nodes. Therefore, the research attempts to achieve the intelligent design of complex traffic node organizations based on digital analogue-driven and integrated building information models and optimization graph convolution algorithms. It aims to achieve real-time monitoring and analysis of traffic information to improve the overall performance of the transportation system. These experiments confirmed that the studied BIM could achieve full lifecycle management of urban interchanges, and could achieve strain analysis and trend prediction based on time series. When the organizational members were 122, the information channels for smart stations based on research technology were 241, which improved information efficiency by more than 90%. When the sensor node was 48, the predicted values of the research model based on the optimization graph convolution algorithm were consistent with the actual values, with a prediction accuracy of 95%. Therefore, the technology studied in this study can provide reliable solutions for the design of complex traffic nodes and support traffic management decisions.

关键词： Complex transportation nodes urban interchange transportation big data BIM

来源：评论

学校读者我要写书评

暂无评论

Optimization and Application of algorithms in Educational Management Information Systems: Improving data processing Efficiency and Decision Support

Optimization and Application of Algorithms in Educational Ma...

引用

Computers, Information processing and Advanced Education (CIPAE), 2020 international conference on

作者： Yannan Wang Sichuan University of Media and Communications Chengdu Sichuan China

ISBN: (数字)9798331527662

ISBN: (纸本)9798331527679

The growth in the volume and complexity of data within educational institutions detail the need for robust methods and systems for effective data collection and processing in making informed decisions. This research paper, therefore, seeks to examine and explore the application and optimization of advanced algorithms in Educational Management Information Systems (EMIS). As established, the paper focuses on strengthening decision-making support systems and enhancing data processing and efficiency. A comprehensive examination of existing algorithms and their inclusion in the educational sector will play a fundamental role in identifying improvement elements and establishing the need for innovative optimization models and techniques. This study, therefore, demonstrates how the optimization of algorithms may be used to reduce data processing periods, thus enhancing the efficiency and accuracy of predictive models that provide actionable insights to different stakeholders. Therefore, this study's findings underscore a transformative measure that may be used in optimizing EMIS, thus paving the way for data-driven and real-time educational management systems.

关键词： Optimization models Electromagnetic interference Decision making Management information systems Predictive models data collection Prediction algorithms Real-time systems Stakeholders Optimization

来源：评论

学校读者我要写书评

暂无评论

CUTRANSNET: TRANSFORMERS TO MAKE STRONG ENCODERS FOR MULTI-TASK VISION PERCEPTION OF AUTONOMOUS DRIVING 49

CUTRANSNET: TRANSFORMERS TO MAKE STRONG ENCODERS FOR MULTI-T...

引用

49th IEEE international conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Li, Jianping Ke, Xiao Wang, Zhihao Wan, JinCheng Tan, Guozhen Dalian Univ Technol Sch Comp Sci & Technol Dalian Peoples R China Fuzhou Univ Coll Comp & Data Sci Fuzhou Peoples R China

ISBN: (纸本)9798350344868;9798350344851

In autonomous driving, perception plays a critical role as it serves as a fundamental requirement for both planning and control. Currently, most perception tasks are processed independently, which requires designing multiple models and networks to handle multiple tasks. This division leads to multiple sub-tasks in real-world environments, making it difficult to ensure real-time performance and communication between tasks. The paper introduces an innovative neural network architecture termed CUTransNet, which presents a unified approach capable of simultaneously detecting drivable areas, lane markings, and traffic objects, achieving multi-task processing with a single model. To build a more robust feature encoder, we proposed the CUT module that combines the global context ability of Transformers. The module is integrated into the convolutional neural network's backbone to compensate for the low-level visual clues lost by Transformers and achieve higher detection accuracy. Experimental results on BDD100K demonstrate that CUT model surpasses traditional multi-task networks in both task accuracy and computational efficiency while maintaining high real-time performance.

关键词： Vision Perception Multi-task Network Transformer Real-Time Performance Feature Encoder

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：