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

709 篇 会议
329 篇 期刊文献
5 册 图书

馆藏范围

1,043 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

642 篇 工学
- 480 篇 计算机科学与技术...
- 405 篇 软件工程
- 147 篇 信息与通信工程
- 85 篇 控制科学与工程
- 66 篇 生物工程
- 51 篇 电气工程
- 47 篇 机械工程
- 42 篇 电子科学与技术（可...
- 42 篇 生物医学工程（可授...
- 36 篇 光学工程
- 32 篇 仪器科学与技术
- 31 篇 化学工程与技术
- 25 篇 交通运输工程
- 23 篇 建筑学
- 21 篇 土木工程
- 20 篇 动力工程及工程热...
299 篇 理学
- 158 篇 数学
- 78 篇 物理学
- 71 篇 生物学
- 49 篇 统计学（可授理学、...
- 31 篇 化学
- 26 篇 系统科学
215 篇 管理学
- 134 篇 管理科学与工程(可...
- 86 篇 图书情报与档案管...
- 62 篇 工商管理
36 篇 医学
- 33 篇 临床医学
- 24 篇 基础医学(可授医学...
- 17 篇 药学(可授医学、理...
25 篇 法学
- 25 篇 社会学
18 篇 经济学
- 18 篇 应用经济学
11 篇 教育学
9 篇 农学
6 篇 文学
3 篇 军事学
1 篇 艺术学

主题

34 篇 training
32 篇 feature extracti...
30 篇 accuracy
27 篇 deep learning
27 篇 software enginee...
24 篇 computer science
22 篇 software
21 篇 semantics
20 篇 machine learning
19 篇 information syst...
18 篇 data mining
16 篇 computational mo...
15 篇 neural networks
15 篇 visualization
14 篇 optimization
14 篇 wireless sensor ...
13 篇 support vector m...
13 篇 databases
13 篇 testing
12 篇 reliability

机构

36 篇 national enginee...
22 篇 college of compu...
20 篇 school of softwa...
16 篇 school of inform...
15 篇 university of ch...
15 篇 guangdong provin...
13 篇 department of co...
12 篇 school of softwa...
12 篇 state key labora...
11 篇 hubei key labora...
11 篇 school of comput...
11 篇 guangdong key la...
11 篇 school of inform...
11 篇 school of cyber ...
10 篇 national enginee...
10 篇 school of comput...
10 篇 shenzhen institu...
10 篇 hubei engineerin...
9 篇 cas center for e...
9 篇 services computi...

作者

24 篇 shen linlin
12 篇 yunhao liu
11 篇 linlin shen
10 篇 zheng changwen
10 篇 herawan tutut
9 篇 hu shengshan
9 篇 qiang wenwen
9 篇 zhang leo yu
8 篇 bin hu
8 篇 wang jinbao
8 篇 gao can
8 篇 li minghui
7 篇 ming gu
7 篇 yu zitong
7 篇 sutoyo edi
7 篇 si lingyu
7 篇 chiroma haruna
6 篇 mungad m.
6 篇 li jiangmeng
6 篇 arshah ruzaini a...

语言

992 篇 英文
29 篇 其他
24 篇 中文

检索条件"机构=Information System and Software Engineering"

共 1043 条记录，以下是21-30 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：

SlotFusion: Object-Centric Audiovisual Feature Fusion with Slot Attention for Remote Sensing Scene Recognition

SlotFusion: Object-Centric Audiovisual Feature Fusion with S...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Han, Fangzhou Yu, Tianyi Zhang, Lamei Si, Lingyu Zhang, Yiqi Dept. of Information Engineering Harbin Institute of Technology Harbin China National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences Beijing China Institute of Software Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798350368741

Despite significant advancements in remote sensing multimodal learning, particularly in image-image feature fusion, the exploration of audio-image feature fusion remains insufficient. Given the complexity and redundancy of ground objects in remote sensing images, accurately aligning audio features with image features during the fusion process is a critical challenge. In this paper, we introduce an object-centric feature fusion method named SlotFusion. By employing a slot attention-based feature decoupling module and a slot-based audiovisual feature fusion module, we transform modality features with complex semantic information into a set of slot features corresponding to object units and use gated activation units to adaptively implement object-centric feature fusion. Experiments on the Audio Visual Aerial Scene Recognition dataset (ADVANCE) demonstrate that the proposed SlotFusion significantly improves remote sensing scene recognition performance, with a 7.04% increase in overall accuracy compared to previous methods, achieving state-of-the-art results. © 2025 IEEE.

关键词： Audiovisual feature fusion Gated feature fusion Object-centric learning Slot attention

来源：评论

学校读者我要写书评

暂无评论

Intelligent Computing in Communication Networks: Challenges, Solutions, and Research Prospects

Intelligent Computing in Communication Networks: Challenges,...

引用

2023 IEEE International Conference on Computing, ICOCO 2023

作者： Fakhrul Safitra, Muhammad Prastika Putri, Deyana Lubis, Muharman Fakhrurroja, Hanif Saidi Lubis, Fahdi Satria, Villy School of Industrial and System Engineering Telkom University Information System Department Bandung Indonesia School of Industrial and System Engineering Telkom University Industrial Engineering Department Bandung Indonesia School of Applied Science Telkom University Application Software Engineering Department Bandung Indonesia

ISBN: (纸本)9798350302684

Communication networks can be transformed through the implementation of computing, which can improve their overall performance, security, and energy efficiency. Nevertheless, several obstacles must be overcome before computing can be fully integrated into communication networks. This paper examines the challenges and research opportunities in the field of computing in communication networks. To take full advantage of computing, new algorithms and techniques must be developed, and machine learning must be explored. Big data analytics and the integration of edge and fog computing with computing techniques are also promising areas for research. The challenges that must be addressed include ethical and privacy concerns, energy consumption, integration with 5G networks, IoT, and cloud computing, and the development of resilient computing systems that can identify and respond to cyber threats. Overcoming these challenges will ensure that computing remains an essential tool for designing and optimizing communication networks. © 2023 IEEE.

关键词： Data Analytics

来源：评论

学校读者我要写书评

暂无评论

Non-Autoregressive Sign Language Translation: A Transformer Encoder-only Approach for Enhanced Multimodal Integration 24

Non-Autoregressive Sign Language Translation: A Transformer ...

引用

2024 International Conference on Advanced Robotics, Automation engineering and Machine Learning, ARAEML 2024

作者： Du, Yao Xie, Pan Peng, Taiying Hu, Xiaohui School of Automation Science and Electrical Engineering Beihang University Beijing China Science and Technology on Integrated Information System Laboratory Institute of Software Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798400717116

Sign language translation aims to transform continuous sign language videos into text, thereby mapping sign language actions onto natural language. Current research primarily focuses on the Transformer's encoder-decoder architecture, employing an autoregressive (AR) approach for SLT. This one-by-one token generation method has a slow speed to translate. We design a Non-Autoregressive Sign Language Translation (NA-SLT) model based on Transformer Encoder. The proposed model consists of three components: The text-encoder, sign-encoder, and sign-Text-encoder. Both the text-encoder and sign-encoder utilize the Encoder to model single-modality data. The sign-Text-encoder effectively engages with multi-modal data, producing high-quality corresponding text after minimal refinement iterations. Moreover, during model training, we devise a double-mask strategy. In addition to traditional random masking, the schedule allocates mask probabilities based on each word's frequency, ensuring long-Tail tokens have an increased chance of being masked and receiving extensive training. We evaluate the model on PHOENIX-2014T, and the results indicate that NA-SLT approaches the performance of AR models on BLEU (21.06) and ROUGE(49.59) metrics while significantly improving translation speed (5.4x-7.9x). © 2024 ACM.

关键词： Metadata

来源：评论

学校读者我要写书评

暂无评论

Optimizing Hate Speech Detection in the Indonesian Language using FastText and LSTM Algorithms 2

Optimizing Hate Speech Detection in the Indonesian Language ...

引用

2nd International Symposium on information Technology and Digital Innovation, ISITDI 2024

作者： Wenando, Febby Apri Yusoff, Nooraini Izrin, Nurul Universitas Andalas Department of Information System Padang Indonesia Universiti Malaysia Kelantan Department of Data Science Kelantan Malaysia Universiti Teknikal Malaysia Department of Software Engineering Malaysia

ISBN: (纸本)9798350389241

The term hate speech encompasses communication that conveys hostility or prejudice toward individuals or groups based on specific attributes, including race, ethnicity, gender, or religion. Given the prevalence of such harmful content on major social media platforms like Twitter, it is imperative to prioritize ongoing research into automated detection methods. This study employs advanced artificial intelligence and machine learning techniques, involving pivotal stages such as data collection, preprocessing, oversampling, fastText word embedding, and classification using Long Short-Term Memory (LSTM) networks to achieve the objective. The amalgamation of fastText embeddings with LSTM classification yielded noteworthy results, with the model achieving 94% accuracy, 92% precision, 94% recall, and a 93% F1 score. These outcomes underscore the effectiveness of sophisticated machine learning methodologies in accurately discerning and addressing the proliferation of hate speech online, particularly in the context of Indonesian-language content. © 2024 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Trusted Polarimetric Feature Fusion for Polsar Image Classification

Trusted Polarimetric Feature Fusion for Polsar Image Classif...

引用

2023 SAR in Big Data Era, BIGSARDATA 2023

作者： Han, Fangzhou Zhang, Lamei Dong, Hongwei Harbin Institute of Technology Department of Information Engineering Harbin150001 China Institute of Software Chinese Academy of Sciences Science & Technology on Integrated Information System Laboratory Beijing China

ISBN: (纸本)9798350326239

Benefiting from the rapid advancements in Deep Learning (DL), data-driven features exhibit remarkable separability in PolSAR image classification. However, data-driven features are challenging to interpret and susceptible to resolution and noise. On the contrary, polarimetric features based on the target decomposition not only have high robustness, but also has physical interpretability. Consequently, researchers have begun to explore ways to integrate both types of features to achieve enhanced performance, which typically concat the data-driven features and polarimetric features directly. However, polarimetric features are not stable in separability, which is because the inherent defects of the target decomposition algorithm. To mitigate the adverse effects of low-quality features, this paper proposes a trusted polarimetric feature fusion (TPFF). The new paradigm calculates the confidence of features and employs a novel feature fusion method. The proposed method has been tested on the EMISAR Foulum dataset, with experimental results demonstrating improved classification performance and robustness. © 2023 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Preliminary denoising by 3D U-Net in image domain for low dose CT images 2022

Preliminary denoising by 3D U-Net in image domain for low do...

引用

2nd International Conference on Bioinformatics and Intelligent Computing, BIC 2022

作者： Song, Xiaofu Han, Yu Xi, Xiaoqi Li, Lei Zhu, Linlin Yang, Shuangzhan Liu, Mengnan Tan, Siyu Yan, Bin Zhengzhou University School of Software Zhengzhou45000 China Information Engineering University College of Information System Engineering Zhengzhou45000 China

ISBN: (纸本)9781450395755

Low dose CT (LDCT) by reducing the X-ray tube current is of huge significance during clinical scanning. However, low-dose CT images often have strong noise and artifacts, which affects the image quality and diagnostic performance. LDCT noise reduction methods based on deep learning have recently achieved good results in improving image quality. Since the reconstructed CT image itself is 3D, in this paper a LDCT denoising method based on 3D U-Net is proposed to combine the 3D spatial information by 3D convolution directly, instead of processing 2D slices from 3D volume data. Therefore, the image change continuity between the adjacent slices is guaranteed. In addition, multiple down-sampling operations in the network, which can reduce the number of parameters of the 3D network, help the network to train. The experimental results show that the proposed method can effectively preserve the structural and texture information of normal NDCT images and significantly suppress the image noise and artifacts, achieving better performance in both quantification and visualization. Compared with LDCT images without denoising, the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) of the processed images were improved by 12.18 dB and 0.35 dB, respectively. © 2022 ACM.

关键词： Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Prioritization Approach by integrating non-Functional and Functional User Stories in Agile-Scrum software Development (i-USPA):A preliminary study

A Hybrid Prioritization Approach by integrating non-Function...

引用

2022 IEEE International Conference on Computing, ICOCO 2022

作者： Borhan, Noor Hazlini Zulzalil, Hazura Hassan, Sa'Adah Ali, Norhayati Mohd Universiti Malaysia Sarawak Department of Software Engineering Sarawak Malaysia Universiti Putra Malaysia Department of Software Engineering & Information System Selangor Malaysia

ISBN: (纸本)9781665489966

Due to its significance in the creation of software projects, the Agile-Scrum methodology has only lately become well-known in the field of software development. The Scrum technique, which is a process that gradually, iteratively, and continuously provides software based on time boxes, is another study that supports agile practitioners' recent shift towards this method (sprints). It consists of user stories that are delivered during sprints by a Scrum team made up of team members, a Scrum Master, and a Product Owner. User stories are kept in product backlogs. The integrated needs prioritization methodologies have been created by a small number of scholars. However, the majority of the research is on non-agile software development. The prioritization of both functional and nonfunctional user stories simultaneously during Agile-Scrum software Development (ASSD) is one of the main gaps that has been ignored by all of these methodologies, according to the literature review conducted by the researchers. The purpose of this study is to outline a research plan for creating an integrated user story prioritizing approach that combines non-functional and functional user stories of the ASSD (i-USPA). The preliminary findings demonstrate the critical importance of understanding the significance of non-functional user stories or requirements during the early stages of software development using the agile methodology, particularly in ASSD, in order to produce high-quality software while staying within budget and time constraints. The data was acquired from a small group of experts or software practitioners who use Agile, particularly the Scrum technique, in their organizations. © 2022 IEEE.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

An improved scattering correction method for model analysis of cone-beam CT 2022

An improved scattering correction method for model analysis ...

引用

2nd International Conference on Bioinformatics and Intelligent Computing, BIC 2022

作者： Kang, Guanyu Han, Yu Xi, Xiaoqi Li, Lei Yang, Shuangzhan Zhu, Linlin Liu, Mengnan Tan, Siyu Yan, Bin Zhengzhou University School of Software Zhengzhou45000 China Information Engineering University College of Information System Engineering Zhengzhou45000 China

ISBN: (纸本)9781450395755

Due to the increased area of X-ray interaction with the objects, the cone beam computed tomography (CBCT) systems will bring severe scattering artifacts in reconstructed images, and the correction of which is quite essential to a good image quality. The scatter process modeling methods without Monte Carlo simulation proposed in recent years is relatively efficient to obtain the scattering distribution. In this paper, combined with Compton scattering cross section (K-N) formula, an improved scatter correction method is proposed for single-material objects. After setting the scattering points in the object from the reconstruction image, it is derived that the scattering ratio of any two scattering points at scattering angle 0 ° is approximately equal to the projection ratio of the object on the corresponding path of two scattering points. Therefore, the first-order Compton scattering values of all scattering points at scattering angle 0 ° can be calculated. The scattering probability of the same scattering point at different scattering angles are then normalized to obtain the scattering ratio. The scattering distribution of each scattering point was calculated by combining the scattering value at 0 ° and the included angle and distance between the scattering point and the detector. Since the attenuation coefficient selection and the attenuation path length calculation before the scattering points are omitted, the scattering estimation error is reduced, and the scattering estimation efficiency is improved. The experimental correction results for the CT images of the aircraft aluminum castings show that the proposed method can effectively restore the structural information of scattering effects, enhancing the contrast to noise ratio (CNR) and reducing the cup shadow from 9.07% to 5.49%. © 2022 ACM.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Deep Learning to Classify Bacterial Species in the same Genus

Deep Learning to Classify Bacterial Species in the same Genu...

引用

2024 International Conference on Electronics, information, and Communication, ICEIC 2024

作者： Sheela, Sherin Piang, May Phu Sakorntanant, Sakda Poomrittigul, Suvit Treebupachatsakul, Treesukon School of Engineering King Mongkut's Institute of Technology Ladkrabang Department of Biomedical Engineering Thailand Pathumwan Institute of Technology Faculty of Science and Technology Department of Software Engineering and Information System Bangkok Thailand School of Information Technology King Mongkut's Institute of Technology Ladkrabang Thailand

ISBN: (纸本)9798350371888

Bacterial strains in the same genus share highly similar morphology, gram-staining characteristics, colony sizes, and spatial arrangements. Therefore, identifying them by deep learning can be quite challenging. This study aimed to assess the classification of 7 species of bacteria from 2 genera of Bacillus and Vibrio by using 8 Convolutional Neural Network (CNN) models. We implemented Python programming along with Keras API within the Jupyter Notebook. The models were constructed and evaluated under unbalanced and balanced datasets by augmentation (rotation, flip, etc.). Transfer learning with fine-tuning, and pre-processing of mixup and label smoothing were also applied to reduce overfitting and enhance generalization. Based on the experimental results on private dataset, the results of InceptionResNetV2 emerged as the top-performing model with a notable accuracy of 82.8%, 88.6% precision, 78.4% recall, and 78.0% F1-score when label smoothing was applied at 0.5 on balanced dataset. © 2024 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Free speech or Free Hate Speech? Analyzing the Proliferation of Hate Speech in Parler 6

Free speech or Free Hate Speech? Analyzing the Proliferation...

引用

6th Workshop on Online Abuse and Harms, WOAH 2022

作者： Israeli, Abraham Tsur, Oren Department of Software and Information System Engineering Ben-Gurion University of the Negev Beer-Sheva Israel

ISBN: (纸本)9781955917841

Social platforms such as Gab and Parler, branded as ‘free-speech’ networks, have seen a significant growth of their user base in recent years. This popularity is mainly attributed to the stricter moderation enforced by mainstream platforms such as Twitter, Facebook, and Reddit. In this work we provide the first large scale analysis of hate-speech on Parler. We experiment with an array of algorithms for hate-speech detection, demonstrating the limitations of transfer learning in that domain, given the illusive and ever changing nature of the ways hate-speech is delivered. In order to improve classification accuracy we annotated 10K Parler posts, which we use to fine-tune a BERT classifier. Classification of individual posts is then leveraged for the classification of millions of users via label propagation over the social network. Classifying users by their propensity to disseminate hate, we find that hate mongers make about 16% of Parler active users, and that they have distinct characteristics comparing to other user groups. We find that hate mongers are more active, more central, express distinct levels of sentiment, and convey a distinct array of emotions like anger and sadness. We further complement our analysis by comparing the trends observed in Parler to those found in Gab. To the best of our knowledge, this is among the first works to analyze hate speech in Parler in a quantitative manner and on the user level. © 2022 Association for Computational Linguistics.

关键词： Classification (of information)

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105页 << < 1 2 3 4 5 6 7 8 9 10 > >>

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：