检索结果-内蒙古大学图书馆

Joint Conference of 59th Annual Meeting of the Association-for-Computational-Linguistics (ACL) / 11th international Joint Conference on Natural Language Processing (IJCNLP) / 6th workshop on Representation Learning for NLP (RepL4NLP)

作者： Sui, Dianbo Tian, Zhengkun Chen, Yubo Liu, Kang Zhao, Jun Univ Chinese Acad Sci Natl Lab Pattern Recognit Inst Automat CASSch Artificial Intelligence Beijing Peoples R China

ISBN: (纸本)9781954085527

In this paper, we aim to explore an uncharted territory, which is Chinese multimodal named entity recognition (NER) with both textual and acoustic contents. To achieve this, we construct a large-scale human-annotated Chinese multimodal NER dataset, named CNERTA. Our corpus totally contains 42,987 annotated sentences accompanying by 71 hours of speech data. based on this dataset, we propose a family of strong and representative baseline models, which can leverage textual features or multimodal features. Upon these baselines, to capture the natural monotonic alignment between the textual modality and the acoustic modality, we further propose a simple multimodal multitask model by introducing a speech-to-text alignment auxiliary task. through extensive experiments, we observe that: (1) Progressive performance boosts as we move from unimodal to multimodal, verifying the necessity of integrating speech clues into Chinese NER. (2) Our proposed model yields state-of-the-art (SoTA) results on CNERTA, demonstrating its effectiveness.

关键词： Speech

来源：评论

学校读者我要写书评

暂无评论

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2013年 7877 LNCS卷 VI页

作者： Kropatsch, Walter G. Artner, Nicole M. Haxhimusa, Yll Jiang, Xiaoyi Vienna University of Technology Department of Pattern Recognition and Image Processing Favoritenstr. 9-11 186-3 1040 Vienna Austria University of Münster Department of Computer Science Einsteinstr. 62 48149 Münster Germany

来源：评论

学校读者我要写书评

暂无评论

Fusion based blind image seganalysis by boosting feature selection

Fusion based blind image seganalysis by boosting feature sel...

引用

6th international workshop on Digital Watermarking, IWDW 2007

作者： Dong, Jing Chen, Xiaochuan Guo, Lei Tan, Tieniu National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences P.O. Box 2728 Beijing 10080 China

ISBN: (纸本)3540922377

In this paper, a feature-level fusion based approach is proposed for blind image steganalysis. We choose three types of typical higher-order statistics as the candidate features for fusion and make use of the Boosting Feature Selection (BFS) algorithm as the fusion tool to select a subset of these candidate features as the new fusion feature vector for blind image steganalysis. Support vector machines are then used as the classifier. Experimental results show that the fusion based approach increases the blind detection accuracy and also provides a good generality by identifying an untrained stego-algorithm. Moreover, we evaluate the performance of our candidate features for fusion by making some analysis of the components of the fusion feature vector in our experiments. © 2008 Springer Berlin Heidelberg.

关键词： Higher order statistics

来源：评论

学校读者我要写书评

暂无评论

Discrete Geometry for Computer Imagery - 16th IAPR international Conference, DGCI 2011, Proceedings

Discrete Geometry for Computer Imagery - 16th IAPR Internati...

引用

16th IAPR international Conference on Discrete Geometry for Computer Imagery, DGCI 2011

ISBN: (纸本)9783642198663

the proceedings contain 43 papers. the topics discussed include: a probabilistic grouping principle to go from pixels to visual structures;an arithmetic and combinatorial approach to three-dimensional discrete lines;smooth 2D coordinate systems on discrete surfaces;an improved Riemannian metric approximation for graph cuts;associating cell complexes to four dimensional digital objects;hierarchic Euclidean skeletons in cubical complexes;a unified topological framework for digital imaging;isthmus-based 6-directional parallel thinning algorithms;quasi-linear transformations, numeration systems and fractals;path-based distance with varying weights and neighborhood sequences;sparse object representations by digital distance functions;efficient robust digital hyperplane fitting with bounded error;circular arc reconstruction of digital contours with chosen Hausdorff error;and an error bounded tangent estimator for digitized elliptic curves.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proceedings Sixth international Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing and First ACIS international workshop on Self-Assembling Wireless Networks SNPD/SAWN 2005

Proceedings Sixth International Conference on Software Engin...

引用

6th international Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing and 1st ACIS international workshop on Self-Assembling Wireless Networks, SNPD/SAWN 2005

ISBN: (纸本)0769522947

the proceedings contain 71 papers. the topics discussed include: soft computing algorithms applied to the segmentation of nerve cell images;pattern recognition based on time-frequency distributions of radar micro-Doppler dynamics;a quantitative software quality evaluation model for the artifacts of component based development;a new approach to software requirements elicitation;using data mining technology to design an intelligent CIM system for IC manufacturing;data mining for imprecise temporal associations;analysis of breast cancer using data mining and statistical techniques;analyzing the conditions of coupling existence based on program slicing and some abstract information-flow;a study of model layers and reflection;a general scalable implementation of fast matrix multiplication algorithms on distributed memory computers;error prediction for multi-classification;an integer support vector machine;and layered neural networks computations.

关键词： Software engineering

来源：评论

学校读者我要写书评

暂无评论

Contextual swarm-based multi-layered lattices: A new architecture for contextual pattern recognition

引用

6th IAPR international workshop on Document Analysis Systems, DAS 2004

作者： Elliman, David G. Youssef, Sherin M. School of Computer Science and IT Nottingham University United Kingdom

ISBN: (纸本)3540230602

this paper introduces a novel approach for dynamic structuring of contextual lattices. It is anticipated that the approach can be applied to improve the accuracy of word-segmentation patterns in autonomous text recognition systems. A multi-level hierarchical structure of lattices is used to implement the algorithm, and the approach can be applied in a generic manner to other pattern recognition problems. We apply a top-down structural model in parallel with a constrained probabilistic model and intelligent distributed searching paradigm. this paradigm is based on the integration between probabilistic bi-grams and adaptive intelligent swarm-based agent search to identify the most likely sentence structures. the searching paradigm allows the exploitation of positive feedback as a search mechanism and, consequently, makes the model amenable to parallel implementation. the distributed intelligence of the proposed approach enables the dynamic structuring of contextual lattices and has proved to scale well with large lattice sizes. Moreover, we believe that the proposed architecture solves the ill-conditioned nature of most pattern recognition problems that lies in the effect of noise in the segmentation phase. To verify the developed Swarm-based Intelligent Search Algorithm (SISA), a simulation study was conducted on a set of variable size scripts. the proposed paradigm proved to be efficient in identifying the most highly segmented patterns and also returned good decisions concerning lower probability segments enabling further resegmentations and re-combinations to take place. the paper is the first to apply the intelligent swarm-based paradigm for the identification of optimal segmented patterns in contextual recognition models. the algorithm is compared with other algorithms for the same problem, and the computational results demonstrate that the proposed approach is very efficient and robust for large-scale statistical contextual-lattice structures. © Springer-Verlag

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Word–wise script identification from indian documents

引用

6th IAPR international workshop on Document Analysis Systems, DAS 2004

作者： Sinha, Suranjit Pal, Umapada Chaudhuri, B.B. Computer Vision and Pattern Recognition Unit Indian Statistical Unit 203 B.T. Road Kolkata700 108 India

ISBN: (纸本)3540230602

In a country like India, a single text line of most of the official documents contains two different script words. Under two-language formula, the Indian documents are written in English and the state official language. For Optical Character recognition (OCR) of such a document page, it is necessary to separate different script words before feeding them to the OCRs of individual scripts. In this paper a robust technique is proposed to extract word-wise script identification from Indian doublet form documents. Here, at first, the document is segmented into lines and then the lines are segmented into words. Using different topological and structural features (like number of loops, headline feature, water reservoir concept based features, profile features, etc.) individual script words are identified from the documents. the proposed scheme is tested on 24210 words of different doublets and we received more than 97% accuracy, on average. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

An ontological service oriented and web searching mechanism in distributed e-learning repositories

An ontological service oriented and web searching mechanism ...

引用

6th international Conference on Advanced Learning Technologies, ICALT 2006

作者： Votis, Konstantinos Alexakos, Christos Giotopoulos, Kostas Likothanassis, Spiridon Pattern Recognition Lab. Dept. of Computer Engineering and Informatics University of Patras Greece

ISBN: (纸本)0769526322

the great heterogeneity of web based Learning systems storing and providing digital e-learning data requires the introduction of interoperability aspects in order to resolve integration problems in a flexible and dynamic way. Learning Objects developed and stored in many different places on the web have a tremendous potential to benefit e-learning in particular and education in general. So the distributed e-learning data integration is significant for the enforcement of novel searching mechanisms in distributed e-learning repositories Our approach introduces an advanced search mechanism which initializes a semantic model representing the digital content stored in distributed e-learning servers, through the use of ontologies In order to retrieve useful content to be added to the learning environment while bypassing compatibility problems between different ontological representations of the same domain, we will use a mixed approach constituting both keywords and ontologies Searching tasks are carried out in the metadata level, where information concerning web digital content is published, managed and stored in the form of a scalable description of knowledge domains. © 2006 IEEE.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Toward Fully Exploiting Heterogeneous Corpus: A Decoupled Named Entity recognition Model with Two-stage Training

<i>Toward Fully Exploiting Heterogeneous Corpus</i>: A Decou...

引用

作者： Hu, Yun Zhu, Yeshuang Zhang, Jinchao Zheng, Changwen Zhou, Jie Chinese Acad Sci Inst Software Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Tencent Inc Pattern Recognit Ctr WeChat Shenzhen Peoples R China

ISBN: (纸本)9781954085541

Named Entity recognition (NER) is a fundamental and widely used task in natural language processing (NLP), which is generally trained on the human-annotated corpus. However, data annotation is costly and time-consuming, which restricts its scale and further leads to the performance bottleneck of NER models. In reality, we can conveniently collect large-scale entity dictionaries and distantly supervised data. However, the collected dictionaries are lack of semantic context and the distantly supervised training instances contain large noise, which will bring uncertain effects to NER models when directly incorporated into the high-quality training set. To address the above issue, we propose a BERT-based decoupled NER model with two-stage training to appropriately take advantage of the heterogeneous corpus, including dictionaries, distantly supervised instances, and human-annotated instances. Our decoupled model consists of a Mention-BERT and a Context-BERT to respectively learn from the context-deficient dictionaries and noised distantly supervised instances at the pre-training stage. At the unifiedtraining stage, the two BERTs are trained together on human-annotated data to predict the correct labels for candidate regions. Empirical studies on three Chinese NER datasets demonstrate that our method achieves significant improvements against several baselines, establishing the new state-of-the-art performance.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

graphics recognition. Ten Years Review and Future Perspectives 1

引用

丛书名： Lecture Notes in Computer Science

1000年

作者： Wenyin Liu Josep Lladós

ISBN: (数字)9783540347125

ISBN: (纸本)9783540347118

this book contains refereed and improved papers presented at the 6th IAPR workshop on graphics recognition (GREC 2005). this year is the tenth anniversary of GREC, which was started in 1995 and has been held every 2 years: GREC 1995 in Penn State University, USA (LNCS Volume 1072, Springer, 1996); GREC 1997 in Nancy, France (LNCS Volume 1389, Springer, 1998); GREC 1999 in Jaipur, India (LNCS Volume 1941, Springer, 2000); GREC 2001 in Kingston, Canada (LNCS Volume 2390, Springer, 2002); and GREC 2003 in Barcelona, Spain (LNCS Volume 3088, Springer, 2004). GREC is the main event of IAPR TC-10 (the Technical Committee on graphics recognition within the international Association for pattern recognition) and provides an excellent opportunity for researchers and practitioners at all levels of experience to meet colleagues and to share new ideas and knowledge about graphics recognition methods. graphics recognition is a particular field in the domain of document analysis, which combines pattern recognition and image processing techniques for the analysis of any kind of graphical information in documents from either paper or electronic formats. In its 10 year history, the graphics recognition community has extended its research topics from the analysis and understanding of graphic documents (including engineering drawings vectorization and recognition), to graphics-based information retrieval and symbol recognition, to new media analysis, and even stepped into research areas of other communities, e. g.

关键词： Image Processing and Computer Vision pattern recognition Computer Applications Computer graphics Artificial Intelligence Discrete Mathematics in Computer Science

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：