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

3,111 篇 会议
20 册 图书
18 篇 期刊文献

馆藏范围

3,149 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

2,113 篇 工学
- 1,871 篇 计算机科学与技术...
- 997 篇 软件工程
- 259 篇 控制科学与工程
- 252 篇 电气工程
- 234 篇 信息与通信工程
- 198 篇 生物工程
- 149 篇 交通运输工程
- 105 篇 生物医学工程（可授...
- 96 篇 机械工程
- 60 篇 光学工程
- 57 篇 电子科学与技术（可...
- 54 篇 安全科学与工程
- 49 篇 材料科学与工程（可...
- 47 篇 动力工程及工程热...
- 41 篇 建筑学
- 36 篇 土木工程
506 篇 理学
- 257 篇 数学
- 204 篇 生物学
- 120 篇 统计学（可授理学、...
- 90 篇 物理学
- 57 篇 系统科学
400 篇 管理学
- 263 篇 管理科学与工程(可...
- 161 篇 图书情报与档案管...
- 99 篇 工商管理
201 篇 教育学
- 198 篇 教育学
165 篇 医学
- 160 篇 临床医学
- 79 篇 基础医学(可授医学...
- 52 篇 药学(可授医学、理...
69 篇 法学
- 58 篇 社会学
32 篇 经济学
- 32 篇 应用经济学
17 篇 农学
7 篇 文学
1 篇 艺术学

主题

293 篇 machine learning
290 篇 deep learning
142 篇 training
125 篇 data models
120 篇 neural networks
114 篇 accuracy
105 篇 feature extracti...
101 篇 data mining
87 篇 predictive model...
83 篇 artificial intel...
82 篇 computational mo...
68 篇 machine learning...
61 篇 students
57 篇 reinforcement le...
56 篇 real-time system...
46 篇 convolutional ne...
45 篇 big data
42 篇 internet of thin...
42 篇 training data
42 篇 humans

机构

30 篇 school of comput...
11 篇 institute of inf...
9 篇 kazan natl res t...
9 篇 school of cyber ...
7 篇 school of comput...
6 篇 school of inform...
5 篇 inesc tec porto
5 篇 hefei univ techn...
5 篇 wroclaw univ sci...
4 篇 univ michigan an...
4 篇 school of automa...
4 篇 chinese acad sci...
4 篇 college of compu...
4 篇 univ minho algor...
4 篇 university of ba...
4 篇 school of mechan...
4 篇 xi an jiao tong ...
4 篇 nanyang technol ...
4 篇 singapore manage...
4 篇 university of mi...

作者

7 篇 yang yang
7 篇 santos matilde
6 篇 peng jun
6 篇 taghi m. khoshgo...
6 篇 wang fei
5 篇 li peng
5 篇 horvath tomas
5 篇 wozniak michal
5 篇 machado jose
5 篇 r.c. luo
4 篇 zhang tao
4 篇 lee seong-whan
4 篇 ksieniewicz pawe...
4 篇 zhang lei
4 篇 yang yi
4 篇 novais paulo
4 篇 li li
4 篇 trong-hop do
4 篇 liu qin
4 篇 shimizu jun

语言

3,083 篇 英文
58 篇 其他
13 篇 中文

检索条件"任意字段=21st International Conference on Intelligent Data Engineering and Automated Learning"

共 3149 条记录，以下是3121-3130 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：

Artificial Intelligence-Driven Algorithms for Precision Forecasting in Electricity Pricing

Artificial Intelligence-Driven Algorithms for Precision Fore...

引用

IEEE India conference (INDICON)

作者： Ashis Pradhan Rekha Pilliadugula Nareddy Nageswara Reddy Jagadish Kumar Bokam Hanumantha Rao Bokkisam Department of Electrical Engineering National Institute of Technology Rourkela Odisha India department of Medicine (Nanomaterials) University of Atacama Chile Department of EEE VNRVJIET Telangana India Department of Electrical Electronics and Communication Engineering GITAM University Andhra Pradesh India

ISBN: (数字)9798350391282

ISBN: (纸本)9798350391299

Electricity price forecasting is essential for ensuring the efficient and cost-effective operation of the power industry. Accurate and precise prediction of electricity prices is critical in the electricity market due to the inherent challenges posed by factors such as high volatility, seasonality, calendar effects, and non-linearity. This paper presents four deep learning-based models—Multilayer Perceptron (MLP), Recurrent Neural Network (RNN), Long Short-Term Memory (LstM), and a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN- LstM)—to forecast electricity prices in the Nord Pool Spot Market. The approach utilizes historical electricity price data, making it a univariate time-series forecasting method. The performance of these models is assessed using various accuracy metrics and visualized through box plots. The simulation results indicate that the CNN-LstM model consistently outperforms the other three models across most evaluation metrics.

关键词： Measurement Solid modeling Accuracy Recurrent neural networks Electricity Pricing Predictive models Convolutional neural networks Forecasting Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Early Diabetic Retinopathy Severity Level Detection Using EffUNets

Early Diabetic Retinopathy Severity Level Detection Using Ef...

引用

IEEE India conference (INDICON)

作者： Umang Tripathi Maninder Singh Amrita Mohan Santosh Kumar Tripathy Autonomy Technologies (M.Sc.) Department Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) Erlangen Germany Department of Electronics and Communication Engineering Greater Noida Institute of Technology Greater Noida Greater Noida INDIA Department of Computer Science and Engineering National Institute of Technology Patna Bihar INDIA

ISBN: (数字)9798350391282

ISBN: (纸本)9798350391299

Diabetic retinopathy (DR), a leading cause of blindness worldwide, arises from long-term diabetes mellitus affecting retinal blood vessels. Its prevalence underscores the critical need for early detection to prevent irreversible vision loss. In this paper, we utilize transfer learning models, particularly EfficientNetB0 and EfficientNetB7, in combination with U-Net to detect and classify DR severity levels in fundus images. The fundus images undergo preprocessing, which includes cropping, resizing, wiener filtering, and CLAHE on the green channel for denoising and image enhancement, respectively. Class imbalance is addressed through weight balancing. Preprocessed images are then processed to extract valuable information for DR detection by extracting features using four different hybrid deep convolutional neural network (CNN) models. Our experiments are conducted on the publicly available APTOS dataset, facilitating a comprehensive comparative analysis of various hybrid model combinations. Results demonstrate that the proposed hybrid architectures outperform traditional transfer learning models in capturing intricate features for DR detection and classification, achieving superior sensitivity, precision, and F1 score. The EffUNetB7 model (combining EfficientNetB7 for downsampling and UNet for upsampling) achieves the highest sensitivity of 91.96, precision of 91.90, and F1-score of 91.77.

关键词： Diabetic retinopathy Analytical models Sensitivity Wiener filters Transfer learning Blindness Feature extraction Retina data mining Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Label Semantics-Filtered Enhanced Prototypical Network for Few-Shot Named Entity Recognition

Label Semantics-Filtered Enhanced Prototypical Network for F...

引用

international conference on Wavelet Active Media Technology and Information Processing (ICWAMTIP)

作者： Xucheng Wu Jia Lian Hui Gao Youbo Wang School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Sichuan College of Architectural Technology Chengdu China

ISBN: (数字)9798331519254

ISBN: (纸本)9798331519261

In the task of Named Entity Recognition (NER), few-shot learning has garnered significant attention due to its potential in scenarios with limited labeled data. Existing research often relies on label names to enhance semantic representation, but the information provided by label names is limited, which affects classification performance. To address this issue, this paper proposes a Label Semantics-filtered Enhanced Prototypical Network (LSE-PN), which integrates rich label annotation information to improve the model's knowledge representation. The model employs a BERT encoder to extract semantic features from both text and labels, and optimizes category differentiation through a semantic fusion module. The prototypical network learns intermediate representations from the support set to achieve category clustering, with the model matching the nearest labels and prototype centers based on distance metrics. Experimental results indicate that LSE-PN outperforms existing methods across multiple few-shot NER benchmark tests. Our model is highly effective in cross-domain scenarios.

关键词： Measurement Annotations Target recognition Computational modeling Semantics Prototypes Named entity recognition Knowledge representation Media Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Solar Energy Forecasting Using Microwave Remote Sensing and Artificial Neural Networks

Solar Energy Forecasting Using Microwave Remote Sensing and ...

引用

IEEE India conference (INDICON)

作者： Deigratia Sutnga Winnie Dkhar Bikramjit Goswami Rakesh Roy Electrical Engineering Department NIT Meghalaya Shillong India Electrical and Electronics Engineering NIT Meghalaya Shillong India Electrical and Electronics Engineering Assam Don Bosco University Guwahati India

ISBN: (数字)9798350391282

ISBN: (纸本)9798350391299

As renewable energy sources become increasingly integrated into the grid, accurate day-ahead solar energy prediction is critical for sound energy planning and grid operations. In the current study, we evaluate the potential of satellite remote sensing data using microwave frequency bands in the prediction of solar energy generation potential of a solar photovoltaic (PV) system two days in advance. These results suggest that the brightness temperature at the 10 GHz frequency has the strongest correlation with the actual solar energy output, and even better when this integrates with several of the other best frequencies such as 18Ghz and 36Ghz, paving the way for good prospects for solar energy predictive modelling levels for a couple of days in the future. In order to enhance these findings, a machine learning methodology based on Artificial Neural Network (ANN) artificial intelligence is employed after an indiscriminate selection of data is filtered to find the most well-suited frequencies. However, applying ANN greatly improves the prediction accuracy and yields very good results. This study demonstrates the potential of microwave remote sensing satellite data and artificial neural networks (ANN) in solar energy production prediction. The life sciences industry faces labor shortages and talent competition, making knowledge transfer between old and young employees critical for sustainable success. The findings pave the way towards a transition to a greener and more sustainable energy future through the efficient management and integration of renewables into the existing power system.

关键词： Renewable energy sources Accuracy Satellites Artificial neural networks Solar energy Power system stability Microwave theory and techniques Frequency estimation Microwave filters Remote sensing

来源：评论

学校读者我要写书评

暂无评论

EEG Signal-Based Sentiment Analysis for Brain-Computer Interfaces: A Comprehensive Comparison Between stacking-Ensemble and Deep learning Approaches

EEG Signal-Based Sentiment Analysis for Brain-Computer Inter...

引用

IEEE India conference (INDICON)

作者： Sapthak Mohajon Turjya Raj Singh Bibhuti Bhusan Dash Mahendra Kumar Gourisaria Sudhansu Shekhar Patra Anjan Bandyopadhyay School of Computer Engineering KIIT Deemed to be University Bhubaneswar India School of Computer Applications KIIT Deemed to be University Bhubaneswar India

ISBN: (数字)9798350391282

ISBN: (纸本)9798350391299

The development of brain-computer interfaces (BCIs) is mainly based on sentiment analysis powered by electroencephalogram (EEG) signals, which offers up revolutionary possibilities in domains such as entertainment, healthcare, and human-computer interaction. The proposed research expedition focuses on EEG signal-based sentiment analysis, with a full comparison of stacking-ensemble and cutting-edge deep learning methods. The EEG data used for training came from the Deap website and was meticulously preprocessed using tasks such as artifact removal, band power calculation, band power feature selection, and dimensionality reduction. The obtained EEG data is utilized to train stacking-ensemble and deep neural network designs. The stacking-ensemble model consisted of a two-layer stacked structure, the first layer consisted of four parallel models which included Decision Tree, Random Forest, K-Nearest Neighbors, and XGBoost, and the second layer consisted of logistic regression as the meta-classifier. The deep neural network architecture was built upon three layers to grasp important information from the preprocessed EEG data and produce sentiment outputs. The proposed methodologies while training were evaluated via cross-validation, and post-training with metrics such as overall accuracy, precision, recall, and F1 scores. Via evaluation, the results depicted that the stacking-ensemble approach outperforms the deep neural network architecture. The proposed research study emphasizes the importance of a stacking-ensemble-based approach for EEG sentiment analysis, and excavating a pathway for future research activities that should focus on enacting strategies for improved BCI technologies.

关键词： Training Deep learning Sentiment analysis Artificial neural networks Medical services Nearest neighbor methods Brain modeling Electroencephalography Brain-computer interfaces Random forests

来源：评论

学校读者我要写书评

暂无评论

Encoding Version History Context for Better Code Representation

Encoding Version History Context for Better Code Representat...

引用

international Mining Software Repositories, MSR

作者： Huy Nguyen Christoph Treud Patanamon Thongtanunam The University of Melbourne Australia Singapore Management University Singapore

ISBN: (数字)9798400705878

ISBN: (纸本)9798350363982

With the exponential growth of AI tools that generate source code, understanding software has become crucial. When developers comprehend a program, they may refer to additional contexts to look for information, e.g. program documentation or historical code versions. Therefore, we argue that encoding this additional contextual information could also benefit code representation for deep learning. Recent papers incorporate contextual data (e.g. call hierarchy) into vector representation to address program comprehension problems. This motivates further studies to explore additional contexts, such as version history, to enhance models’ understanding of programs. That is, insights from version history enable recognition of patterns in code evolution over time, recurring issues, and the effectiveness of past solutions. Our paper presents preliminary evidence of the potential benefit of encoding contextual information from the version history to predict code clones and perform code classification. We experiment with two representative deep learning models, AstNN and CodeBERT, to investigate whether combining additional contexts with different aggregations may benefit downstream activities. The experimental result affirms the positive impact of combining version history into source code representation in all scenarios; however, to ensure the technique performs consistently, we need to conduct a holistic investigation on a larger code base using different combinations of contexts, aggregation, and models. Therefore, we propose a research agenda aimed at exploring various aspects of encoding additional context to improve code representation and its optimal utilisation in specific *** Concepts•Computing methodologies → Neural networks; • Software and its engineering → Reusability.

关键词： Deep learning Codes Source coding Neural networks Documentation Software Vectors

来源：评论

学校读者我要写书评

暂无评论

Hiding Digital Watermark Based on Wavelet Decomposition

Hiding Digital Watermark Based on Wavelet Decomposition

引用

ICIG’2000 First international conference on Image and Graphics--Image and Graphic Technology Toward 21st Century and Beyond

作者： Hua LI Guangxi ZHU Yaoting ZHU state Education Commission Laboratory for Image Processing and intelligent Control, Dept. of Electronics and Information engineering, Huazhong University of Science and Technology, 430074 Wuhan,China

Digital watermarking is motivated by the need to provide copyright protection to multimedia data and is a favorable, prospective method of copyright protection. In this paper, a new method is proposed to hide watermark image in images by multi-resolution wavelet decomposition techniques in conjunction with human visual model. The watermark image is processed as a visually recognizable pattern-binary image, which includes more information than the traditional symbol or ID number and is "extracted" instead of only "detected" to characterize the owner. Firstly, the human vision system(HVS) model and wavelet-based decomposition are introduced, part of decomposition coefficients which are applicable to hide watermark are also selected;then, both the watermark and the host image are decomposed into multi-resolution sub-band images, the sub-band watermarks is embedded into the corresponding sub-band host images;finally the experimental results and conclusion are given. The simulation experiments demonstrate the high robustness of the algorithm to cropping and compression, show that this approach has excellent performance in the quality of watermarked image and the robustness of watermark.

关键词： Digital watermark, HVS model, DWT transform, Radial frequency

来源：评论

学校读者我要写书评

暂无评论

Item Analysis of an Instrument to Measure Chemistry students' Critical Thinking Skills Based on Science, Technology, engineering and Mathematics (stEM) on Solubility and Solubility Product (KSP) material using the Rasch model 2020

Item Analysis of an Instrument to Measure Chemistry Students...

引用

Proceedings of the 4th international conference on learning Innovation and Quality Education

作者： Agustin Kumala Dewi Sentot Budi Rahardjo Suryadi Budi Utomo Master Program of Chemistry Education Sebelas Maret University Surakarta Indonesia Master of Chemistry Education Postgraduate Sebelas Maret University Surakarta Indonesia

ISBN: (纸本)9781450375726

In stEM-based learning students use science, technology, engineering, and mathematics in real interactions that connect schools, the world of work, and the global world to develop stEM literacy that enables students to be able to compete in the 21st century. The stEM learning approach by integrating its four components is able to produce students 'thinking activities that are useful to help raise students' critsical thinking which is characterized by the ability to solve problems, make decisions, analyze assumptions, evaluate, and conduct investigations. Evaluation tools have an important role in the learning process. The purpose of this study was to analyze critical thinking skill items on solubility and solubility product (Ksp) material using the Rasch model in terms of validity, reliability, and item difficulty level. The method used is descriptive quantitative with 106 participants students from three High School in Surakarta. data collection techniques in this study are documentation. The number of items as many as 30 questions obtained results, namely the validity of the suitability level of items that are fit as many as 30 questions. student reliability 0.44 is weak, item reliability 0.97 is good and reliability between students with item 0.47 is sufficient. The level of difficulty of the questions is very difficult, difficult and easy is dominated by questions categorized as difficult. In conclusion, the analysis of critical thinking skill items on solubility and solubility product (Ksp) material using the Rasch model is valid, reliable, and the level of difficulty items was very good. So the questions that have been analyzed can be used to determine critical thinking skills.

关键词： Item Analysis stEM rasch model solubility and solubility product Critical Thinking Skills

来源：评论

学校读者我要写书评

暂无评论

ViZig: Anchor Points based Non-Linear Navigation and Summarization in Educational Videos 16

ViZig: Anchor Points based Non-Linear Navigation and Summari...

引用

Proceedings of the 21st international conference on intelligent User Interfaces

作者： Kuldeep Yadav Ankit Gandhi Arijit Biswas Kundan Shrivastava Saurabh Srivastava Om Deshmukh Xerox Research Centre India Bangalore Karnataka India Xerox Research Center India Bangalore Karnataka India Xerox Research Center Bangalore Karnataka India Xerox Research Centre India Bangalore Karnataka India Xerox Research Center India Bangalore Karnataka India

ISBN: (纸本)9781450341370

Instructional videos are one of the most popular ways of teaching and learning in an online setting. However, navigation in videos is linear as compared to other instructional resources such as textbooks, where a table of topics and a multi-faceted index of different anchor points i.e., list of figures, tables aid in efficiently navigating to a desired point of interest. There is a lack of appropriate techniques and interfaces which can support such textbook-style navigation in instructional videos. This paper presents a novel approach to automatically localize and classify different anchor points in a video including figures, tables, equations, flowcharts, code snippets and charts. Our approach uses a deep convolution neural network in a semi-supervised fashion where the training data is obtained from the unconstrained Internet images. On an anchor point dataset of about 10K images, the proposed algorithm leads to a classification accuracy of 86%. Further, we designed a system ViZig that uses these localized anchor points along with a automatically generated list of topics for non-linear video navigation and studied its effectiveness in real-world. Our user studies with 18 participants establish that the proposed video navigation mechanism provides statistically significant time savings as compared to the popularly used time-synched transcript along with youtube-style timeline scrubbing.

关键词： video indexing video summarization moocs deep learning education video navigation convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Exploring Reproducibility of fMRI Analysis Using SPM and Nilearn Open-Source Tools

Exploring Reproducibility of fMRI Analysis Using SPM and Nil...

引用

international conference on Electrical engineering, Computing Science and Automatic Control (CCE)

作者： Héctor G. Martínez-Fuentes Mario I. Gutierrez Jorge A. Mercado-Gutierrez Josefina Gutiérrez-Martínez Fisica Biomedica UMAM Facultad de Ciencias Mexico City Mexico Division de Inv. en Ing. Medica Inst. Nac. de Rehabilitacion “LGII” Mexico City Mexico

ISBN: (数字)9798350377545

ISBN: (纸本)9798350377552

Functional magnetic resonance imaging (fMRI) is a widely used technique for studying human brain activity. Various libraries and tools are available for performing the analysis and preprocessing of data. However, a significant challenge is that these different tools often yield varying results, which can complicate the interpretation and comparison of studies. In this context to contribute to the transparency of the used framework and reproducibility of the results in fMRI studies, this work presents a comparative analysis using Nilearn, a library available for Python, and SPM, a graphical interface available in MATLAB, along with preprocessing conducted with SPM in Nipype, also available for Python. Blood oxygen level- dependent (BOLD) contrast and anatomical volumes from one subject of the Flanker Task dataset were used. For this purpose, three distinct contrasts were established: Congruent > Incongruent, Congruent > Baseline, and Incongruent > Baseline. A point-by-point comparison of these masks was performed using a density plot and visualization of the different masks in the subject's anatomical volume. The results indicated that the design matrices of SPM and Nilearn are slightly different despite being configured with the same parameters, leading to significant differences in the final results.

关键词： Computational modeling Closed box Machine learning Functional magnetic resonance imaging Electrical engineering computing Reproducibility of results Libraries MATLAB Python Programming environments

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315页 << < 306 307 308 309 310 311 312 313 314 315 > >>

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：