Query：
学者姓名：于娟
Refining：
Year
Type
Indexed by
Source
Complex
Co-
Language
Clean All
Abstract ：
Corporations require screening critical information from numerous resumes with different formats and content for managerial decision-making. However, traditional manual screening methods have low accuracy to meet the demand. Therefore, we propose a multimodal network model incorporating entity semantic graphs, ESGNet, for accurately extracting critical informa-tion from Chinese resumes. Firstly, each resume is partitioned into distinct blocks according to content while constructing an entity semantic graph according to entity categories. Then we interact with associated features within image and text modalities to capture the latent semantic information. Furthermore, we employ Transformer containing multimodal self-attention to establish relationships among modalities and incorporate supervised comparative learning concepts into the loss function for categorizing feature information. The experimental results on the real Chinese resume dataset demonstrate that ESGNet achieves the best information extraction results on all three indicators compared with other models, with the comprehensive indicator F1 score reaching 91.65%.
Keyword ：
Deep learning Deep learning Entity semantic graphs Entity semantic graphs Multimodal network model Multimodal network model Resume information extraction Resume information extraction
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Luo, Shun , Yu, Juan . ESGNet: A multimodal network model incorporating entity semantic graphs for information extraction from Chinese resumes [J]. | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (1) . |
| MLA | Luo, Shun 等. "ESGNet: A multimodal network model incorporating entity semantic graphs for information extraction from Chinese resumes" . | INFORMATION PROCESSING & MANAGEMENT 61 . 1 (2024) . |
| APA | Luo, Shun , Yu, Juan . ESGNet: A multimodal network model incorporating entity semantic graphs for information extraction from Chinese resumes . | INFORMATION PROCESSING & MANAGEMENT , 2024 , 61 (1) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
跨语言文本主题发现是跨语言文本挖掘领域的重要研究方向，对跨语言文本分析和组织各种文本数据具有较高的应用价值。基于Bagging和跨语言词嵌入改进LDA主题模型，提出跨语言文本主题发现方法BCL-LDA(Bagging, Cross-lingual word embedding with LDA),从多语言文本中挖掘关键信息。该方法首先将Bagging集成学习思想与LDA主题模型结合生成混合语言子主题集；然后利用跨语言词嵌入和K-means算法对混合子主题进行聚类分组；最后使用TF-IDF算法对主题词进行过滤排序。汉语-德语、汉语-法语主题发现实验表明，该方法在主题连贯性和多样性方面均表现优异，能够提取出语义更加相关且主题更加连贯多样的双语主题。
Keyword ：
LDA LDA 主题发现 主题发现 主题聚类 主题聚类 德语 德语 法语 法语 跨语言 跨语言
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 李帅 , 于娟 , 巫邵诚 . 基于集成学习的跨语言文本主题发现方法研究 [J]. | 计算机科学 , 2024 , 51 (S1) : 194-201 . |
| MLA | 李帅 等. "基于集成学习的跨语言文本主题发现方法研究" . | 计算机科学 51 . S1 (2024) : 194-201 . |
| APA | 李帅 , 于娟 , 巫邵诚 . 基于集成学习的跨语言文本主题发现方法研究 . | 计算机科学 , 2024 , 51 (S1) , 194-201 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对现有虚假信息检测方法主要基于单模态数据分析,检测时忽视了信息之间相关性的问题,提出了结合社交网络图的多模态虚假信息检测模型.该模型使用预训练Transformer模型和图像描述模型分别从多角度提取各模态数据的语义,并通过融合信息传播过程中的社交网络图,在文本和图像模态中加入传播信息的特征,最后使用跨模态注意力机制分配各模态信息权重以进行虚假信息检测.在推特和微博两个真实数据集上进行对比实验,所提模型的虚假信息检测准确率稳定为约88％,高于EANN、PTCA等现有基线模型.实验结果表明所提模型能够有效融合多模态信息,从而提高虚假信息检测的准确率.
Keyword ：
多模态融合 多模态融合 社交网络图 社交网络图 网络舆情 网络舆情 虚假信息检测 虚假信息检测 跨模态注意力 跨模态注意力
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 叶舟波 , 罗舜 , 于娟 . 结合社交网络图的多模态虚假信息检测模型 [J]. | 计算机应用研究 , 2024 , 41 (7) : 1992-1998 . |
| MLA | 叶舟波 等. "结合社交网络图的多模态虚假信息检测模型" . | 计算机应用研究 41 . 7 (2024) : 1992-1998 . |
| APA | 叶舟波 , 罗舜 , 于娟 . 结合社交网络图的多模态虚假信息检测模型 . | 计算机应用研究 , 2024 , 41 (7) , 1992-1998 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Documents contain abundant information available for managerial decision-making. However, manual methods of screening document information lack accuracy due to the heterogeneity of documents. To address the above issue, we propose a multimodal network combining multivariate semantic association graphs, MMIE, for accurately extracting information from documents. Firstly, the multivariate semantic graphs between multimodal data within each modality are constructed based on the semantic association of text contents, followed by the semantic relationships in the graphs to lead the fusion and embedding of the extracted multimodal data and improve the feature representation capability. Subsequently, the semantically linked multimodal information is fed into the newly constructed multimodal self-attention module to better establish inter-modal associations. Finally, a supervised comparison learning loss function is employed to reduce further the information loss due to sample imbalance. The experimental results on three real datasets show that the proposed model can extract feature information of different modal data more accurately, and the F1 scores reach 87.28%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document}, 82.53%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document}, and 81.17%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document} , respectively.
Keyword ：
Deep learning Deep learning Document information extraction Document information extraction Multimodal fusion Multimodal fusion Multivariate semantic association Multivariate semantic association
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Luo, Shun , Yu, Juan , Xi, Yunjiang . Incorporating multivariate semantic association graphs into multimodal networks for information extraction from documents [J]. | JOURNAL OF SUPERCOMPUTING , 2024 , 80 (13) : 18705-18727 . |
| MLA | Luo, Shun 等. "Incorporating multivariate semantic association graphs into multimodal networks for information extraction from documents" . | JOURNAL OF SUPERCOMPUTING 80 . 13 (2024) : 18705-18727 . |
| APA | Luo, Shun , Yu, Juan , Xi, Yunjiang . Incorporating multivariate semantic association graphs into multimodal networks for information extraction from documents . | JOURNAL OF SUPERCOMPUTING , 2024 , 80 (13) , 18705-18727 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
[目的/意义]提出一种适用于德语文本处理的文本相似度计算方法，填补了国内外德语文本聚类研究的空缺。[方法/过程]通过词语提取和特征词选择将每个德语文本表示为一个特征词的集合，寻找集合间配对的特征词对，由特征词对的匹配度得到文本间的相似度。[结果/结论]基于多个德语数据集的实验结果表明，相比于已有方法，本文提出的基于特征词配对的德语文本聚类方法提升了约5%的NMI值和约6%的Purity值。基于特征词配对的相似度计算方法能够保留更多的文本信息，从而进一步提升德语文本聚类的性能。
Keyword ：
德语 德语 文本相似度 文本相似度 文本聚类 文本聚类 特征词 特征词
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 简梓炜 , 于娟 . 基于特征词配对的德语文本聚类方法研究 [J]. | 情报探索 , 2022 , 8 (09) : 86-93 . |
| MLA | 简梓炜 等. "基于特征词配对的德语文本聚类方法研究" . | 情报探索 8 . 09 (2022) : 86-93 . |
| APA | 简梓炜 , 于娟 . 基于特征词配对的德语文本聚类方法研究 . | 情报探索 , 2022 , 8 (09) , 86-93 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Purpose: This article aims to provide an integrated model and methodology for discovering valuable innovation knowledge and creative users in online innovative communities. Design/methodology/approach: The super-network integration approach is applied in constructing the user's innovation knowledge super-network (UIKSN) model based on knowledge fragments discovered from user-generated content (UGC) data using text mining methods. The social network analysis (SNA) methodology is then used to analyze the KSN model. An empirical research is conducted with China's Xiaomi online community to illustrate the UIKSN model and the SNA methodology. Findings: Compared with current methods, the proposed KSN model and SNA methodology are demonstrated to be more effective and valid in discovering valuable innovation knowledge including valuable innovations, innovation trends, creative users and their knowledge background, etc. Research limitations/implications: The results of KSN modeling and analysis are inevitably influenced by the performance of text mining methods. Practical implications: More and more companies start to adopt online community user's innovations in their new product developments. However, it is difficult to discover users' opinions and innovations from the UGC due to its tremendous volume. Therefore, the KSN model and SNA methodology presented in this article are helpful for companies to manage online communities and to utilize users' innovations effectively and efficiently. Originality/value: The article provides two main contributions in the study of online communities: 1) a well-justified model (users' innovation KSN, UIKSN) to explore online users' contributions; 2) a new and more effective methodology to discover and analyze valuable innovations and innovative users.
Keyword ：
Analytical models Analytical models Companies Companies Integrated model Integrated model Knowledge based systems Knowledge based systems Knowledge engineering Knowledge engineering super-network super-network Technological innovation Technological innovation Text mining Text mining Tools Tools user innovation knowledge user innovation knowledge
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zantow, Kenneth , Yu, Juan , Ye, Guangyu et al. An Integrated Model for User Innovation Knowledge Based on Super-Network [J]. | IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT , 2022 , 69 (2) : 399-408 . |
| MLA | Zantow, Kenneth et al. "An Integrated Model for User Innovation Knowledge Based on Super-Network" . | IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT 69 . 2 (2022) : 399-408 . |
| APA | Zantow, Kenneth , Yu, Juan , Ye, Guangyu , Xi, Yunjiang , Liao, Xiao . An Integrated Model for User Innovation Knowledge Based on Super-Network . | IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT , 2022 , 69 (2) , 399-408 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对自动驾驶情景下行人目标检测过程中对于重叠和遮挡目标存在的漏检问题，提出一种改进多尺度网络YOLOv5的行人目标检测算法.首先，构建同时考虑通道间关系和特征空间位置信息的多重协调注意力模块，增加网络特征表达能力；然后，将原损失函数改进为具有双重惩罚项的切比雪夫距离交并比损失函数，提高检测框的精确度与网络收敛速度；最后，在网络结构方面设计瓶颈状DSP1＿X和DSP2＿X模块减少梯度混淆.实验结果表明，改进后的多尺度网络收敛能力提高，在面对行车中复杂行人目标检测时具有较高的判别精度和实时检测速度.
Keyword ：
YOLOv5 YOLOv5 图像处理 图像处理 损失函数 损失函数 改进多尺度网络 改进多尺度网络 注意力机制 注意力机制 行人目标检测 行人目标检测
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 罗舜 , 于娟 . 改进多尺度网络的行人目标检测算法 [J]. | 福州大学学报(自然科学版) , 2022 , 50 (05) : 587-594 . |
| MLA | 罗舜 et al. "改进多尺度网络的行人目标检测算法" . | 福州大学学报(自然科学版) 50 . 05 (2022) : 587-594 . |
| APA | 罗舜 , 于娟 . 改进多尺度网络的行人目标检测算法 . | 福州大学学报(自然科学版) , 2022 , 50 (05) , 587-594 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Dealing with the insufficient detection accuracy and speed of aircraft targets in remote sensing images under complex background, this paper proposes a new detection method, YOLOv5-Aircraft, based on the YOLOv5 network. The YOLOv5-Aircraft model is improved in 3 ways: (1) At the beginning and end of original batch normalization module, centering and scaling calibration are added to enhance the effective features and form a more stable feature distribution, which strengthens the feature extraction ability of network model. (2) The cross-entropy loss function in the confidence of the original loss function is improved to the loss function based on smoothed Kullback-Leibler divergence. (3) For reducing information loss, the CSandGlass module is designed on the backbone feature extraction network of YOLOv5 to replace the residual module. Meanwhile, low-resolution feature layers are eliminated to reduce semantic loss. Experiment results demonstrate that the YOLOv5-Aircraft model can enhance the accuracy and speed of aircraft target detection in remote sensing images while achieving easier convergence.
Keyword ：
Aircraft Aircraft aircraft detection aircraft detection batch normalization batch normalization Calibration Calibration Convolutional neural networks Convolutional neural networks Feature extraction Feature extraction loss function loss function Object detection Object detection Real-time systems Real-time systems Remote sensing Remote sensing Remote sensing image Remote sensing image YOLOv5 YOLOv5
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Luo, Shun , Yu, Juan , Xi, Yunjiang et al. Aircraft Target Detection in Remote Sensing Images Based on Improved YOLOv5 [J]. | IEEE ACCESS , 2022 , 10 : 5184-5192 . |
| MLA | Luo, Shun et al. "Aircraft Target Detection in Remote Sensing Images Based on Improved YOLOv5" . | IEEE ACCESS 10 (2022) : 5184-5192 . |
| APA | Luo, Shun , Yu, Juan , Xi, Yunjiang , Liao, Xiao . Aircraft Target Detection in Remote Sensing Images Based on Improved YOLOv5 . | IEEE ACCESS , 2022 , 10 , 5184-5192 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
跨语言术语对齐是跨语言文本数据分析与知识发现的关键基础.针对跨语言术语对齐研究多为单词术语对齐且严重依赖向量空间对齐的现状,提出一种能够实现跨语言单词及多词术语间一对多对齐的Kernel-XGBoost方法.给定跨语言平行语料库,该方法分两步得到同义的跨语言术语对:1)跨语言术语提取与候选术语对生成;2)基于跨语言词嵌入的术语对齐.汉语-西班牙语以及汉语-法语的术语对齐实验表明,该方法在Top-5的准确率可达到80％,能有效支持跨语言信息检索、本体构建等跨语言文本数据挖掘任务.
Keyword ：
Kernel-XGBoost Kernel-XGBoost 文本分析 文本分析 术语对齐 术语对齐 汉语 汉语 法语 法语 西班牙语 西班牙语 跨语言 跨语言
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 于娟 , 张晨 . 基于Kernel-XGBoost的跨语言术语对齐方法 [J]. | 计算机科学 , 2022 , 49 (z2) : 114-119 . |
| MLA | 于娟 et al. "基于Kernel-XGBoost的跨语言术语对齐方法" . | 计算机科学 49 . z2 (2022) : 114-119 . |
| APA | 于娟 , 张晨 . 基于Kernel-XGBoost的跨语言术语对齐方法 . | 计算机科学 , 2022 , 49 (z2) , 114-119 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
法语复杂的语法和词形变化规则导致N-gram等词语提取方法的效果无法保证,影响法语文本挖掘的准确性。该文提出一种高效的法文词语提取方法,从待分析的法语文本中自动获取包括单词和短语的词语集合,构建法语文本挖掘所需的词库。该方法把文本中的单词共现信息压缩为FP序列树结构,快速提取频繁词串并计算其成词度,得到法文词语集合。实验表明,该方法的准确率高达90%,且具有比现有法文词语提取方法更高的召回率,能有效支持法语文本挖掘应用。
Keyword ：
FP序列树 FP序列树 成词度 成词度 文本压缩 文本压缩 法语文本挖掘 法语文本挖掘 词语提取 词语提取
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 于娟 , 吴晓鹏 , 廖晓 et al. 基于FP序列树的法文词语提取方法研究 [J]. | 电子科技大学学报 , 2021 , 50 (01) : 84-90 . |
| MLA | 于娟 et al. "基于FP序列树的法文词语提取方法研究" . | 电子科技大学学报 50 . 01 (2021) : 84-90 . |
| APA | 于娟 , 吴晓鹏 , 廖晓 , 刘建国 . 基于FP序列树的法文词语提取方法研究 . | 电子科技大学学报 , 2021 , 50 (01) , 84-90 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
