检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

69,864 篇 会议
40,148 篇 期刊文献
801 册 图书

馆藏范围

110,813 篇 电子文献
1 种 纸本馆藏

日期分布

学科分类号

68,312 篇 工学
- 45,731 篇 计算机科学与技术...
- 37,592 篇 软件工程
- 17,042 篇 信息与通信工程
- 11,243 篇 控制科学与工程
- 8,821 篇 电气工程
- 8,349 篇 电子科学与技术（可...
- 7,911 篇 生物工程
- 5,721 篇 光学工程
- 5,524 篇 生物医学工程（可授...
- 5,439 篇 机械工程
- 3,877 篇 化学工程与技术
- 3,823 篇 仪器科学与技术
- 3,541 篇 动力工程及工程热...
- 2,780 篇 网络空间安全
- 2,685 篇 交通运输工程
- 2,611 篇 材料科学与工程（可...
- 2,559 篇 安全科学与工程
37,875 篇 理学
- 19,106 篇 数学
- 11,734 篇 物理学
- 8,870 篇 生物学
- 6,391 篇 统计学（可授理学、...
- 4,160 篇 化学
- 4,023 篇 系统科学
18,318 篇 管理学
- 11,638 篇 管理科学与工程(可...
- 7,414 篇 图书情报与档案管...
- 4,743 篇 工商管理
5,318 篇 医学
- 4,343 篇 临床医学
- 3,916 篇 基础医学(可授医学...
- 2,363 篇 药学(可授医学、理...
2,947 篇 法学
- 2,401 篇 社会学
1,855 篇 经济学
1,430 篇 农学
1,032 篇 教育学
467 篇 文学
255 篇 军事学
211 篇 艺术学
7 篇 哲学
7 篇 历史学

主题

3,143 篇 accuracy
3,014 篇 feature extracti...
2,796 篇 deep learning
2,612 篇 training
2,420 篇 computational mo...
2,140 篇 computer science
2,008 篇 machine learning
1,644 篇 optimization
1,553 篇 predictive model...
1,409 篇 semantics
1,372 篇 convolutional ne...
1,344 篇 internet of thin...
1,340 篇 real-time system...
1,304 篇 support vector m...
1,290 篇 data models
1,186 篇 neural networks
1,183 篇 data mining
1,038 篇 image segmentati...
999 篇 artificial intel...
926 篇 security

机构

2,477 篇 school of comput...
1,556 篇 school of comput...
1,391 篇 school of comput...
1,086 篇 cavendish labora...
1,017 篇 department of ph...
983 篇 school of comput...
944 篇 physikalisches i...
942 篇 oliver lodge lab...
822 篇 university of ch...
693 篇 school of comput...
572 篇 school of comput...
536 篇 horia hulubei na...
535 篇 fakultät physik ...
534 篇 chitkara univers...
529 篇 school of physic...
528 篇 school of physic...
522 篇 department of ph...
516 篇 imperial college...
515 篇 school of optica...
514 篇 h.h. wills physi...

作者

275 篇 niyato dusit
268 篇 a. gomes
260 篇 barter w.
249 篇 c. alexa
247 篇 j. m. izen
246 篇 g. bella
246 篇 j. strandberg
244 篇 d. calvet
243 篇 c. amelung
238 篇 beiter a.
233 篇 s. chen
231 篇 blake t.
225 篇 bellee v.
225 篇 bowcock t.j.v.
221 篇 betancourt c.
220 篇 amato s.
219 篇 brundu d.
219 篇 borsato m.
218 篇 boettcher t.
217 篇 g. spigo

语言

101,561 篇 英文
7,389 篇 其他
1,834 篇 中文
129 篇 日文
13 篇 德文
11 篇 法文
10 篇 朝鲜文
7 篇 斯洛文尼亚文
2 篇 荷兰文
1 篇 西班牙文

检索条件"机构=School of Computer Science and Engineering Technology"

共 110813 条记录，以下是1131-1140 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Understanding Training Behaviour during Transfer Learning by drawing parallels to Grokking 4

Understanding Training Behaviour during Transfer Learning by...

引用

4th International Conference on Intelligent Technologies, CONIT 2024

作者： Rath, Kanishk Gnanasigamani, Lydia J Vellore Institue of Technology School of Computer Science and Engineering Vellore India

ISBN: (纸本)9798350349900

In the realm of machine learning, understanding how models arrive at their predictions or decisions, known as mechanistic interpretability, is crucial for trust, transparency, and refinement. This involves unraveling the relationships between input features, intermediate representations, and output predictions. One key aspect of mechanistic interpretability is the analysis of attention weights and neural activations in neural network models, which provide insights into the decision-making processes and utilization of learned features. Transfer learning, on the other hand, is a widely used technique in machine learning that involves pre-training a model on one task or dataset and fine-tuning it for another task or dataset. By capitalizing on the transferability of features learned from one domain to another, transfer learning enhances model performance, particularly in scenarios where labeled data is limited. This research paper aims to investigate behavioral patterns trained with and without transfer learning, observing patterns of training, and comparing it with known phenomenon of grokking, we can hypothesize the efficacy of transfer learning during different phases of grokking, through this investigation, the paper aims to contribute to a deeper understanding of transfer learning and its implications for neural network training, with potential implications for improving model performance and real-world applicability. © 2024 IEEE.

关键词： Grokking Mechanistic Interpretability Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

Crafting Trust with Convolutional Neural Networks and Hyperparameter Tuning for Precision Signature Verification in Insurance Claim

Crafting Trust with Convolutional Neural Networks and Hyperp...

引用

2024 International Conference on Advances in Data engineering and Intelligent Computing Systems, ADICS 2024

作者： Balaji, Yashwant Vikas Kedar, Rohit Virupakshi, Hrishikesh Anandan, P. School Of Computer Science And Engineering Vellore Institute Of Technology Chennai India

ISBN: (纸本)9798350364828

The prevention of signature forgery on physical documents is a critical concern in sectors like insurance, where the authenticity of claims documentation is pivotal. Handwritten signatures pose a challenge for verification systems due to inherent variability arising from diverse writing styles, pen types, paper qualities, and deliberate forgery attempts. To enhance the precision and resilience of signature verification in the context of insurance claims, researchers have explored various methods, with Convolutional Neural Networks (CNNs) emerging as a particularly promising approach. Therefore, this research paper aims to compare different CNN models in signature forgery detection, providing insights into their performance, strengths, and weaknesses, and contributing to the advancement of signature verification methods. Through this exploration, we aspire to lay the groundwork for the development of more sophisticated and accurate signature verification methods tailored to the unique challenges posed by insurance claim documentation. Ultimately, this research aims to fortify the security measures in place, ensuring the authenticity of insurance claims and mitigating the risks associated with fraudulent manipulations. © 2024 IEEE.

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

STOE-YOLOv5s: A Hard Hat Wearing Detection Algorithm Based on Improved YOLOv5s 6

STOE-YOLOv5s: A Hard Hat Wearing Detection Algorithm Based o...

引用

6th International Conference on Robotics and computer Vision, ICRCV 2024

作者： Hu, Tao Huang, Wei Wei, Jianghao Liu, Qiang Huang, Jiahuan Yu, Jie School of Computer Science and Engineering Wuhan Institute of Technology Wuhan China

ISBN: (纸本)9798331527426

To address the existing issues of missed detection and false detection of hard hats by mainstream detection models in complex environments, an improved hard hat detection algorithm, STOE-YOLOv5s based on YOLOv5s, has been proposed. Firstly, enhance the detection header structure, add multi-scale target detection layers, and improve the detection capability for small targets;Secondly, a BottleNeck Transformer (BoT3) module based on self-attention is added to the backbone network to enhance the model's ability to extract global features;Then, to solve the problem of positive and negative sample matching, a Simplified Optimal Transport Assignment (SimOT A) strategy is used to obtain the optimal matching result in the global range;Finally, an Efficient Multi-Scale Attention (EMA) mechanism is fused into the C3 module of the network neck to enhance the model's ability to capture features-rich areas. Experimental results on the hard hat dataset show that the mAP of the improved algorithm reaches 96.0%, and the recall rate reaches 91.8%. Compared with the original YOLOv5s algorithm, it has increased by 1.7 and 2.9 percentage points respectively, proving the effectiveness and advancement of the improved algorithm. © 2024 IEEE.

关键词： EMA Hard Hat Wearing Detection SimOTA YOLO

来源：评论

学校读者我要写书评

暂无评论

PSR-Tree: A Novel Method for Personalized Trajectory Data Protection 22

PSR-Tree: A Novel Method for Personalized Trajectory Data Pr...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Wang, Chundong Song, Yixuan Zhao, Yongxin Tian, Yuhan Fang, Shunyao Tianjin University of Technology School of Computer Science and Engineering Tianjin China

ISBN: (纸本)9798331509712

Privacy concerns related to location-based services are becoming increasingly critical. To address the limitations of previous research, which has often neglected the temporal attributes of trajectories and relied solely on average noise addition methods, a trajectory data publishing scheme is proposed based on an R-tree storage structure. Initially, trajectory data is dimensionalized using a spatiotemporal optimal trajectory segmentation algorithm, and the preprocessed key trajectory data is organized into a PSR-Tree. Subsequently, an incremental privacy budget allocation strategy is applied to add noise to the trajectory data. Finally, a Hidden Markov Model is employed to constrain noise addition at the PSR-Tree nodes, fulfilling the requirements for trajectory data publishing. Experimental comparisons with existing methods demonstrate that this approach effectively addresses the issue of insufficient data usability and exhibits robust scalability. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Low-Latency State Management for Real-Time Tasks in Edge Serverless 22

Low-Latency State Management for Real-Time Tasks in Edge Ser...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Wen, Yanyan Xu, Guangping Wang, Jianshe Hao, Wei Tianjin University of Technology School of Computer Science and Engineering Tianjin China

ISBN: (纸本)9798331509712

Stateful serverless systems commonly adopt an architectural paradigm characterized by compute and storage separation within cloud data centers. Nevertheless, guaranteeing prompt response for real-time tasks at the edge becomes challenging due to network overheads. This paper introduces a Low-Latency state management framework for real-time tasks in edge serverless systems called LoLa, which adaptively places states proximate to functions, thereby mitigating delays in accessing states within edge serverless systems. Our approach aims at mitigating network latency and optimizing resource utilization by co-locating functions and states, thereby enhancing the system's overall efficiency. We introduce an adaptive strategy to coordinate the migration of states. It dynamically adjusts the positions of states based on historical data and real-time feedback. Additionally, we designed an in-memory state storage mechanism to facilitate low-latency access and implement a lightweight and fine-grained state management to ensure stored state consistency. Evaluation results showcase the efficacy of LoLa in reducing state read and write latency within edge serverless systems. Specifically, the average response latency is observed to decrease by 65.2% and 38.1% in the best and worst-case scenarios, respectively. © 2024 IEEE.

关键词： Edge computing Stateful Serverless

来源：评论

学校读者我要写书评

暂无评论

Implementing OTP Verification and Zero Trust Security for Role-Based Access Control in Medical Records 3

Implementing OTP Verification and Zero Trust Security for Ro...

引用

3rd International Conference on Smart Technologies and Systems for Next Generation Computing, ICSTSN 2024

作者： Harshini, B.V. Ulagarchana, U. Mounika, Perla Mary Shamala, L. Vellore Institute of Technology School of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350391565

This paper proposes a comprehensive approach for enhancing security in medical record systems by incorporating One-Time Password (OTP) verification and Zero Trust Security principles within a Role-Based Access Control (RBAC) framework. With the increasing digitization of medical records and the growing threat of cyberattacks, ensuring confidentiality, integrity, and availability of patient information is paramount. OTP verification adds a layer of authentication, mitigating the risk of unauthorized access even in the event of compromised credentials. Integrating Zero Trust Security principles ensures that access permissions are continuously evaluated and verified based on various contextual factors, such as device health, user behavior, and network location, rather than relying solely on static credentials. Furthermore, RBAC facilitates granular access control, allowing administrators to assign roles and permissions based on users' responsibilities and organizational hierarchy. By combining OTP verification, Zero Trust Security, and RBAC, healthcare organizations can establish a robust security framework that minimizes the risk of data breaches and ensures compliance with regulatory requirements such as HIPAA. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Alzheimer's Disease Classification Using CUDA-Accelerated DenseNet201 and VGG16 5

Alzheimer's Disease Classification Using CUDA-Accelerated De...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Munugala, Shravya Marathe, Atharva Prashant Vijayakumar, Rahul Kumaran, K. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331518097

Automated classification of Alzheimer's disease through MRI analysis presents significant computational challenges due to complex deep learning architectures and largescale image processing requirements. This paper introduces a GPU-accelerated multi-branch neural network framework that combines modified DenseNet201 and VGG16 architectures for efficient Alzheimer's disease stage classification. The proposed model employs parallel feature extraction pathways, with each branch independently processing MRI inputs before feature fusion. CUDA optimization techniques are implemented across the processing pipeline, including data loading, augmentation, and model training phases, resulting in significantly reduced computational overhead. The framework classifies brain MRI scans into four categories: Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented. Experimental results demonstrate that the proposed hybrid architecture achieves 98% classification accuracy while substantially reducing training time through GPU acceleration. The model's performance is further enhanced through adaptive learning rate scheduling and selective layer fine-tuning. This work establishes an efficient approach for deploying complex neural architectures in clinical neuroimaging applications where computational resources and processing time are critical constraints. © 2024 IEEE.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

An Integrated Framework for Real-Time Disaster Data Acquisition, Processing, and Resource Optimization in Crisis Management Systems 4

An Integrated Framework for Real-Time Disaster Data Acquisit...

引用

4th International Conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Saranya, G. Atchaya, M. Harshini, R. Choudhary, Kartik School of Computer Science Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331529635

The rapid onset and unpredictable nature of disasters necessitate a quick, organized, and efficient response. This paper proposes a comprehensive disaster management platform that integrates real-time disaster information, weather forecasting, and resource management to enhance disaster preparedness and response. The platform collects disaster data from geophysical monitoring stations, processes the data for consistency, and stores it in a MongoDB database. Geospatial visualization tools, such as Folium, are employed to present disaster events on interactive maps, helping decision-makers assess the situation and respond accordingly. Additionally, the platform includes features for weather forecasting and resource management, allowing users to monitor conditions and contribute essential resources in real time. This research demonstrates how leveraging modern technologies can improve disaster management systems, providing a more effective and coordinated response to natural and man-made disasters. © 2024 IEEE.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Optimising Makespan Minimization and Performance at the Same Time with a Dynamic Cloud Task Scheduling Comparative Study of MOSOS and PSO 5

Optimising Makespan Minimization and Performance at the Same...

引用

5th IEEE International Conference on Computing, Power, and Communication Technologies, IC2PCT 2024

作者： Jain, Anegha Nath, Megha Rama Parvathy, L. Vellore Institute of Technology School of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350383522

The practice of allocating available resources, such as personnel, money, time, or materials, to different tasks, projects, or activities is known as resource allocation. Effective resource allocation is crucial for optimizing productivity, achieving goals, and maximizing efficiency within an organization or project. Resource allocation in cloud computing can face several challenges. These include over-allocation, where resources are provisioned excessively, leading to increased cost, and under-allocation, which can result in performance bottlenecks. Cloud environments often have to support a variety of different workloads, each with its own resource requirements. This can make it difficult to allocate resources efficiently whilst maintaining the Quality of Service (QoS). Tackling such issues, our paper demonstrates the workings of the MOSOS (multi-objective symbiotic organism search) algorithm to optimize the performance and cut down on makespan and cost. We simulate MOSOS alongside the Particle Swarm Optimization Algorithm (PSO) to understand whether PSO outperforms MOSOS or whether MOSOS accomplishes similar results to PSO. The results back up MOSOS in terms of makespan but fail when compared against PSO's execution times. We find that PSO uses a lesser number of parameters and makes fewer function calls as opposed to MOSOS. Thus, we can argue that in dynamic cloud environments where time is essential, PSO may perform better as compared to MOSOS. © 2024 IEEE.

关键词： Cloud computing Evolutionary algorithms makespan optimization resource scheduling

来源：评论

学校读者我要写书评

暂无评论

Attention-Enhanced Feature Extraction for Image Defogging Algorithms 6

Attention-Enhanced Feature Extraction for Image Defogging Al...

引用

6th International Conference on Robotics and computer Vision, ICRCV 2024

作者： Yang, Shuwei Xu, Junwu Deng, Zifan School of Computer Science and Engineering Wuhan Institute of Technology Wuhan China

ISBN: (纸本)9798331527426

The AOD-Net suffers from issues such as low image quality and color distortion after defogging. An improved defogging algorithm is proposed, which applies pooling operations and dilated convolution to the convolutional layers of AOD-Net to enhance feature extraction. Multi-scale feature fusion is used, and the extracted information features are fed back to improve propagation, accelerate model training, and enhance the model's generalization ability. Additionally, an ECA attention module is introduced, allowing the network to dynamically adjust the focus on different feature channels, enhancing the network's feature modeling ability and suppressing redundant information, thus improving the defogging quality of the model. Experiments are conducted using the public dataset NYU2 and existing lightweight defogging algorithms for comparison. The defogging quality has improved compared to defogging algorithms such as DCP and CAP. Compared with AOD-Net, the peak signal-to-noise ratio of this method has increased by 4.037 dB, and the structural similarity has reached 0.918, effectively enhancing the network's defogging capability. © 2024 IEEE.

关键词： Attention Module Deep Learning Dilated Convolution Image Defogging

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 110 111 112 113 114 115 116 117 118 119 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：