检索结果-内蒙古大学图书馆

38th Annual International No-Dig Conference and Exhibition 2022, ISTT 2022

作者： Choo, Chung Siung Ong, Dominic Ek Leong Phangkawira, Fredrik Yeo, Lit Yen Lee, Sue Han Faculty of Engineering Computing and Science Swinburne University of Technology Sarawak Campus Sarawak Malaysia School of Engineering and Built Environment Griffith University Nathan Campus QLD Australia WSP Golder Jakarta Indonesia

The estimation of frictional jacking forces is not well understood for pipe jacking drives traversing weathered lithological units, such as those found in the Tuang Formation underlying the central business district (CBD) of Kuching, Malaysia. Pipe jacking was the preferred method for the construction of trunk sewerage pipelines buried at depths of up to 35 m below the busy CBD of Kuching. Many existing jacking force models rely on the strength parameters of the geomaterials negotiated during pipe jacking, with minimal understanding of the influence of pipe jacking operation parameters on jacking forces. The operation parameters during pipe jacking can also be influential towards the accrual of frictional jacking forces. However, one of the key challenges was the extraction of intact rock cores for strength characterization. Therefore, this paper presents some methods of developing equivalent rock strength parameters through the use of shear box testing of reconstituted rock spoil and in situ pressuremeter testing. The outcomes were then used for assessing jacking forces for a drive traversing weathered phyllite. Subsequently, deep learning techniques together with attention mechanism were applied to a drive to visualize the significance of input features (i.e. pipe jacking operation parameters) on jacking forces. © 38th Annual International No-Dig Conference and Exhibition 2022, ISTT 2022.

关键词： Lithology

来源：评论

学校读者我要写书评

暂无评论

Multi-Granular Multimodal Clue Fusion for Meme Understanding

arXiv

引用

arXiv 2025年

作者： Zheng, Li Fei, Hao Dai, Ting Peng, Zuquan Li, Fei Ma, Huisheng Teng, Chong Ji, Donghong Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University Wuhan China National University of Singapore Singapore Singapore Laboratory for Advanced Computing and Intelligence Engineering Wuxi China North China Institute of Computing Technology Beijing China

With the continuous emergence of various social media platforms frequently used in daily life, the multimodal meme understanding (MMU) task has been garnering increasing attention. MMU aims to explore and comprehend the meanings of memes from various perspectives by performing tasks such as metaphor recognition, sentiment analysis, intention detection, and offensiveness detection. Despite making progress, limitations persist due to the loss of fine-grained metaphorical visual clue and the neglect of multimodal text-image weak correlation. To overcome these limitations, we propose a multi-granular multimodal clue fusion model (MGMCF) to advance MMU. Firstly, we design an object-level semantic mining module to extract object-level image feature clues, achieving fine-grained feature clue extraction and enhancing the model’s ability to capture metaphorical details and semantics. Secondly, we propose a brand-new global-local cross-modal interaction model to address the weak correlation between text and images. This model facilitates effective interaction between global multimodal contextual clues and local unimodal feature clues, strengthening their representations through a bidirectional cross-modal attention mechanism. Finally, we devise a dual-semantic guided training strategy to enhance the model’s understanding and alignment of multimodal representations in the semantic space. Experiments conducted on the widely-used MET-MEME bilingual dataset demonstrate significant improvements over state-of-the-art baselines. Specifically, there is an 8.14% increase in precision for offensiveness detection task, and respective accuracy enhancements of 3.53%, 3.89%, and 3.52% for metaphor recognition, sentiment analysis, and intention detection tasks. These results, underpinned by in-depth analyses, underscore the effectiveness and potential of our approach for advancing MMU. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Brain Depletion Recognition Through Iot Sensors Empowered with Computational Intelligence 4th

Brain Depletion Recognition Through Iot Sensors Empowered wi...

引用

4th International Conference on Futuristic Trends in Networks and computing Technologies, FTNCT 2021

作者： Theodore, Shaji. K. A. Selvakumar, K. Revathy, G. Faculty of IT–Networking Department of IT University of Technology and Applied Sciences Al-Masnaah Oman Department of IT Faculty of Engineering and Technology Annamalai University Chidmabaram India School of Computing SASTRA Deemed to Be University Thanjavur India

ISBN: (纸本)9789811950360

The world has been inspired by the development of Internet of Things enabling applications in recent years, which have brought outstanding solutions to a number of problems. This fast-growing industry is led by wireless sensor networks, radio frequency identification, and smart mobile technologies. Smart medical devices and wearables, for example, play an important part in the Internet of Things, as they may collect a variety of longitudinal patient-generated health data while also presenting preliminary diagnosis options. As part of their efforts to support patients utilizing IoT-based solutions, experts exploit the capabilities of machine learning algorithms to provide effective solutions in bleeding detection. This chapter outlines a smart IoT-based system for human brain haemorrhage diagnostics that employs deep learning algorithms to lower fatality rates and provide accurate treatment suggestions. To classify the pictures from the computed tomography scans for the intracranial dataset, the Nave Bayes Classification and Recurrent Neural Network were utilized. The classification results for the Naive Bayes classification and Recurrent neural network are superior to prior techniques such as support vector machines, knn, and kmediods. According to the findings, the recurrent neural network outperforms at classifying intracranial imagery. The output of the classification tool includes information on the type of brain haemorrhage, which aids in the validation of an expert’s diagnosis and is utilised as a learning tool for trainee radiologists to eliminate errors in current systems. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

A Mutual Intrusion Detection System for MANET to Address DDoS attacks using Data Mining Techniques 2

A Mutual Intrusion Detection System for MANET to Address DDo...

引用

2nd International Conference on Sustainable computing and Data Communication Systems, ICSCDS 2023

作者： Bala, K. Renjith, P.N. Latha, M. Santhanalakshmi, S.T. School of Computing Bharath Institute of Higher Education and Research Department of Computer Science and Engineering Tamilnadu Chennai600073 India Department of Computer Science and Engineering VIT Chennai Campus Tamilnadu Chennai600127 India Srm Institute of Science and Technology Department of Information Technology Ramapuram Campus Tamilnadu Chennai600089 India Panimalar Engineering College Department of Computer Science and Engineering Tamil Nadu Chennai600123 India

ISBN: (纸本)9781665491990

A type of wireless communication network called a mobile ad hoc network (MANET) is composed of numerous flexible components that can move in any of the network's many directions. The Intrusion Detection System (IDS) is a systematic tool implemented to spot suspicious activity with network malfunctions and sends an alert to the security control center. A malicious effort to interfere with the server's or networks regular operation is acknowledged as a distributed denial-of-service (DDoS). This occurrence causes the networking system's general performance to decline. Certain measures must be put into place to improve efficiency. These DDoS attacks are classified as application layer attacks, protocol attacks, and volumetric attacks in the networking system. The server experiences downtime as a result, and the network's general performance suffers. When resources are brought back into the network, it causes external tension. Data mining methods can be used to find this and fix it. © 2023 IEEE.

关键词： Mobile ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Tracking Control for a Composite Mobile Robot Considering System Uncertainties

Hierarchical Tracking Control for a Composite Mobile Robot C...

引用

International Conference on Computer and Automation engineering, ICCAE

作者： Yiming Yan Fusen Guo Huadong Mo Xueting Huang School of Mechanical Engineering Southeast University Nanjing China School of Science Computing and Engineering Technologies Swinburne University of Technology Melbourne Australia School of Engineering and Information Technology University of New South Wales Canberra Australia Faculty of Art design Guangzhou College of Commerce Guangzhou China

ISBN: (数字)9798350370058

ISBN: (纸本)9798350370164

Composite mobile robots (CMRs) have gained extensive application in the industrial sector due to their integrated loading and operational capabilities. The trajectory tracking control module, being a vital constituent, directly influences the degree of stability in robot operations. However, current tracking control methods exhibit suboptimal performance when encountering system uncertainties arising from both the environment and inherent system factors. This paper presents a novel hierarchical control strategy tailored for the hierarchical model of composite mobile robot chassis. This strategy effectively combines velocity control and torque control, resulting in satisfactory trajectory tracking control. Firstly, a hierarchical kinematic and dynamic model is constructed to account for varying payload and environmental uncertainties. Subsequently, a pose-velocity-torque hierarchical controller is formulated to regulate chassis speed by manipulating torque input, thereby achieving pose regulation and trajectory tracking. Finally, the control algorithm is implemented using MATLAB and adapted to industrial computers for the development and validation of the controller's effectiveness. Comparative experiments are conducted against existing methods, demonstrating the superiority of the proposed controller.

关键词： Computers Adaptation models Uncertainty Trajectory tracking Torque control Kinematics Mathematical models

来源：评论

学校读者我要写书评

暂无评论

Expediting Distributed GNN Training with Feature-only Partition and Optimized Communication Planning

Expediting Distributed GNN Training with Feature-only Partit...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE Computer and Communications Societies

作者： Bingqian Du Jun Liu Ziyue Luo Chuan Wu Qiankun Zhang Hai Jin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Department of Electrical and Computer Engineering The Ohio State University USA Department of Computer Science The University of Hong Kong Hong Kong School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

Feature-only partition of large graph data in distributed Graph Neural Network (GNN) training offers advantages over commonly adopted graph structure partition, such as minimal graph preprocessing cost and elimination of cross-worker subgraph sampling burdens. Nonetheless, performance bottleneck of GNN training with feature-only partitions still largely lies in the substantial communication overhead due to cross-worker feature fetching. To reduce the communication overhead and expedite distributed training, we first investigate and answer two key questions on convergence behaviors of GNN model in feature-partition based distribute GNN training: 1) As no worker holds a complete copy of each feature, can gradient exchange among workers compensate for the information loss due to incomplete local features? 2) If the answer to the first question is negative, is feature fetching in every training iteration of the GNN model necessary to ensure model convergence? Based on our theoretical findings on these questions, we derive an optimal communication plan that decides the frequency for feature fetching during the training process, taking into account bandwidth levels among workers and striking a balance between model loss and training time. Extensive evaluation demonstrates consistent results with our theoretical analysis, and the effectiveness of our proposed design.

关键词： Training Time-frequency analysis Costs Computational modeling Distributed databases Bandwidth Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision

Molecular Data Programming: Towards Molecule Pseudo-labeling...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Xin Juan Kaixiong Zhou Ninghao Liu Tianlong Chen Xin Wang School of Artificial Intelligence Jilin University China Institute for Medical Engineering & Science Massachusetts Institute of Technology USA School of Computing University of Georgia USA CSAIL Massachusetts Institute of Technology USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

The premise for the great advancement of molecular machine learning is dependent on a considerable amount of labeled data. In many real-world scenarios, the labeled molecules are limited in quantity or laborious to derive. Recent pseudo-labeling methods are usually designed based on a single domain knowledge, thereby failing to understand the comprehensive molecular configurations and limiting their adaptability to generalize across diverse biochemical context. To this end, we introduce an innovative paradigm for dealing with the molecule pseudo-labeling, named as Molecular Data Programming (MDP). In particular, we adopt systematic supervision sources via crafting multiple graph labeling functions, which covers various molecular structural knowledge of graph kernels, molecular fingerprints, and topological features. Each of them creates an uncertain and biased labels for the unlabeled molecules. To address the decision conflicts among the diverse pseudo-labels, we design a label synchronizer to differentiably model confidences and correlations between the labeling functions, which yields probabilistic molecular labels to adapt for specific applications. These probabilistic molecular labels are used to train a molecular classifier for improving its generalization capability. On eight bench-mark datasets, we empirically demonstrate the effectiveness of MDP on the weakly supervised molecule classification tasks, achieving an average improvement of 9.5%. The code is in: https://***/xinjuan1/MDP/.

关键词： Systematics Correlation Limiting Supervised learning Machine learning Programming Probabilistic logic

来源：评论

学校读者我要写书评

暂无评论

A Powerful Algorithm for e-Commerce Credit Risk Analysis

A Powerful Algorithm for e-Commerce Credit Risk Analysis

引用

2023 International Conference on Data Science and Network Security, ICDSNS 2023

作者： Dayanand Lal, N. Guruswamy, Sandhya Lalitha, Vuyyuru Awasthy, Mohan Pareek, Piyush Kumar Gitam University Bengaluru Campus Dept. of Cse Gst Bengaluru India School of Computing and Information Technology Reva University India Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Vaadeswaram India Bharati Vidyapeeth Peeth Deemed to Be University Department of Engineering and Technology Navi Mumbai India Nitte Meenakshi Institute of Technology Department of Ai and Ml Bengaluru India

ISBN: (纸本)9798350301595

It is critical to build a system and model for analysing the credit risk associated with e-commerce to decrease the chance of credit defaults occurring during online transactions. This approach examines a client's income, assets, negative transaction rate, reputation, and third-party credit evaluation findings to determine creditworthiness. The risk assessment index-based credit risk analysis model is built using the RB-XGBoost approach. The credit risk analysis algorithm's training process employs a data balancing technique to increase the system's capacity to forecast accurately. This method is designed to solve the problem of class imbalance in machine learning datasets. The training set is separated into groups at this point, and each group is compared to a specified threshold. The dataset is then made more representative by combining oversampling and undersampling. The system's capacity to analyse the credit risk presented by online shops is improved by balancing data subsets. Online retailers might utilise the proposed model and indexing approach to better assess their clients' creditworthiness and thereby reduce the number of consumers whose payments are late. By suggesting a unique way to address class imbalance in machine learning datasets, this study adds to the expanding body of work on credit risk assessment in online transactions. © 2023 IEEE.

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

Automatic Adaptation Rule Optimization via Large Language Models

arXiv

引用

arXiv 2024年

作者： Ishimizu, Yusei Li, Jialong Xu, Jinglue Cai, Jinyu Iba, Hitoshi Tei, Kenji School of Computing Tokyo Institute of Technology Tokyo Japan Department of Computer Science and Engineering Waseda University Tokyo Japan Graduate School of Information Science and Technology University of Tokyo Tokyo Japan

Rule-based adaptation is a foundational approach to self-adaptation, characterized by its human readability and rapid response. However, building high-performance and robust adaptation rules is often a challenge because it essentially involves searching the optimal design in a complex (variables) space. In response, this paper attempt to employ large language models (LLMs) as a optimizer to construct and optimize adaptation rules, leveraging the common sense and reasoning capabilities inherent in LLMs. Preliminary experiments conducted in SWIM have validated the effectiveness and limitation of our method. Copyright © 2024, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Integrating Large Language Models for Automated Vulnerability Scanning and Reporting in Network Hosts

Integrating Large Language Models for Automated Vulnerabilit...

引用

International Research Conference on Smart computing and Systems engineering (SCSE)

作者： M. Tharuka Sandaruwan Janaka Wijayanayake Janaka Senanayake Department of Industrial Management University of Kelaniya Kelaniya Sri Lanka School of Computing Engineering & Technology Robert Gordon University Aberdeen United Kingdom

ISBN: (数字)9798331523268

ISBN: (纸本)9798331523275

This research explores integrating Large Language Models (LLMs) like GPT-4 and Claude 3.5 into cybersecurity vulnerability scanning to enhance automation and effectiveness. Current tools’ reliance on manual updates and human expertise is highlighted. A literature review identified effective modular architectures and Retrieval-Augmented Generation (RAG) systems for grounding LLMs with cybersecurity knowledge.A Proof of Concept (PoC) tool, developed in Python and tested on the Metasploitable system, evaluated three LLM implementations: GPT-4 Omni, GPT-4 Omni with RAG, and Claude 3.5 Sonnet. The results showed GPT-4 Omni outperformed Claude 3.5, with RAG significantly improving performance. The tool achieved 80% accuracy in identifying and resolving *** study underscores the potential of LLMs to revolutionize vulnerability scanning, making advanced cybersecurity more accessible and effective. Future work should address limitations, enable interactive sessions, create new exploits, and tackle more complex challenges.

关键词： Automation Accuracy Grounding Large language models Retrieval augmented generation Manuals Computer architecture Systems engineering and theory Computer security Systematic literature review

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：