检索结果-内蒙古大学图书馆

IEEE International Conference on Granular computing (GRC)

作者： Hong Hu Zhongzhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences China

Zadeh proposed that there are three basic concepts that underlie human cognition: granulation, organization and causation and a granule being a clump of points (objects) drawn together by indistinguishability, similarity, proximity or functionality. In this paper, we give out a novel definition of Granular computing which can be easily treated by neural network. Perception learning as granular computing tries to study the machine learning from perception information sampling to dimensional reduction and samples classification in a granular way, and can be summaries as two kind approaches:(1) covering learning, (2) svm kind learning. We proved that although there are tremendous algorithms for dimensional reduction and information transformation, their ability can't transcend wavelet kind nested layered granular computing which are very easy for neural network processing.

关键词： Machine learning Neural networks Cognition Computer networks information processing Humans Fuzzy logic Laboratories Learning systems Sampling methods

来源：评论

学校读者我要写书评

暂无评论

Active Learning of Instance-Level Constraints for Semi-supervised Document Clustering

Active Learning of Instance-Level Constraints for Semi-super...

引用

IEEE WIC ACM International Conference on Web Intelligence (WI)

作者： Weizhong Zhao Qing He Huifang Ma Zhongzhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

This paper presents a framework that actively selects informative documents pairs for semi-supervised document clustering. The semi-supervised document clustering algorithm is a Constrained DBSCAN (Cons-DBSCAN), which incorporates instance-level constraints to guide the clustering process in DBSCAN. By obtaining user feedbacks, our proposed active learning algorithm can get informative instance level constraints to aid clustering process. Experimental results show that Cons-DBSCAN with the proposed active learning approach can provide an appealing clustering performance.

关键词： intelligent agent Clustering algorithms Feedback Machine learning Conferences information processing Computers Data mining Learning systems

来源：评论

学校读者我要写书评

暂无评论

Improving tree-to-tree translation with packed forests 09

Improving tree-to-tree translation with packed forests

引用

Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 2 - Volume 2

作者： Yang Liu Yajuan Lü Qun Liu Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781932432466

Current tree-to-tree models suffer from parsing errors as they usually use only 1-best parses for rule extraction and decoding. We instead propose a forest-based tree-to-tree model that uses packed forests. The model is based on a probabilistic synchronous tree substitution grammar (STSG), which can be learned from aligned forest pairs automatically. The decoder finds ways of decomposing trees in the source forest into elementary trees using the source projection of STSG while building target forest in parallel. Comparable to the state-of-the-art phrase-based system Moses, using packed forests in tree-to-tree translation results in a significant absolute improvement of 3.6 BLEU points over using 1-best trees.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Biological-inspired computational modeling of eye-motion control for object detection

Biological-inspired computational modeling of eye-motion con...

引用

Asian Control Conference

作者： Jun Miao Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

ISBN: (纸本)9781424454402

Eye movement plays an important role in human vision system. How to control eye or gaze movement automatically for image understanding is an interesting issue. This paper presents a progress of our research on biological-inspired computational modeling of eye-motion control for object detection in images. The model simulates the single and population cell coding mechanisms for learning visual context and controlling the eye movement. A comparative experiment with three coding systems is carried out and experimental results show the gradual-scale population coding system performs better than the other two coding systems on the average for object detection.

关键词： Computational modeling Biological control systems Object detection Humans Automatic control Biological system modeling Biological information theory Image coding Context modeling Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Joint decoding with multiple translation models 09

Joint decoding with multiple translation models

引用

Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 2 - Volume 2

作者： Yang Liu Haitao Mi Yang Feng Qun Liu Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781932432466

Current SMT systems usually decode with single translation models and cannot benefit from the strengths of other models in decoding phase. We instead propose joint decoding, a method that combines multiple translation models in one decoder. Our joint decoder draws connections among multiple models by integrating the translation hypergraphs they produce individually. Therefore, one model can share translations and even derivations with other models. Comparable to the state-of-the-art system combination technique, joint decoding achieves an absolute improvement of 1.5 BLEU points over individual decoding.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image features optimizing for content-based image retrieval

Image features optimizing for content-based image retrieval

引用

IEEE International Conference on intelligent computing and intelligent Systems (ICIS)

作者： Zhiping Shi Xi Liu Qing He Zhongzhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

ISBN: (纸本)9781424447374;9781424447541

Developing low-dimensional semantics-sensitive features is crucial for content-based image retrieval (CBIR). In this paper, we present a method called M2CLDA (merging 2-class linear discriminant analysis) to capture low-dimensional optimal discriminative features in the projection space. M2CLDA calculates discriminant vectors with respect to each class in the one-vs-all classification scenario and then merges all the discriminant vectors to form a projection matrix. The dimensionality of the M2CLDA space fits in with the number of classes involved. Moreover, when a new class is added, the new M2CLDA space can be approximated by only calculating a new discriminant vector for the new class. The features in the M2CLDA space have better semantic discrimination than those in traditional LDA space. Our experiments show that the proposed approach improves the performance of image retrieval and image classification dramatically.

关键词： Image retrieval Content based retrieval Linear discriminant analysis information retrieval Scattering Vectors Image classification Data mining Principal component analysis Helium

来源：评论

学校读者我要写书评

暂无评论

Automatically Organize Web Text Resources with Frequent Term Tree

Automatically Organize Web Text Resources with Frequent Term...

引用

International Conference on Computer and information technology (CIT)

作者： Xiaofeng Wang Zhongzhi Shi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Graduate University of the Chinese Academy of Sciences Chinese Academy and Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy and Sciences Beijing China

With the expansion of the Web, automatically organizing large scale text resources, e.g. Web pages, becomes very important. Many Web sites, like Google and Yahoo, use hierarchical classification trees to organize text resources in Web. User can easily find the text resources that meet their requirements by navigating these hierarchical classification trees. Typically, the text resources in Web are manually assigned to the nodes of the hierarchical classification tree. This limits the hierarchical classification tree to organize large scale text resources. In this paper, we propose a Frequent Term Tree to improve the ability of hierarchical classification tree in organizing large scale text resources in Web. Different from the Fp-tree which is utilized to efficiently discover frequent patterns, the frequent term tree is used to organize resources with frequent pattern based classification. The frequent term tree can accurately assign text resources to each node of classification tree and improve the ability in organizing resources with the incremental classified text resources. The evaluation of the frequent term tree demonstrates that frequent term tree can effectively and efficiently organize text resources.

关键词： Classification tree analysis Large-scale systems Organizing Navigation Humans information processing Computers information technology Web pages Web sites

来源：评论

学校读者我要写书评

暂无评论

Identifying protein-protein interaction sites using adapted Bayesian classifier

Identifying protein-protein interaction sites using adapted ...

引用

2009 Second ISECS International Colloquium on computing, Communication, Control, and Management, CCCM 2009

作者： Chishe, Wang Jie, Song Fangping, Li Junsong, Lv School of Information Technology Jinling Institute of Technology Nanjing 211169 China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei 230039 China Center of Network University of Science and Technology of China Hefei 230029 China Anhui Statistics Bureau Hefei 230001 China

ISBN: (纸本)9781424442461

Identifying protein-protein interaction sites have important connotations ranging from rational drug design to analysis metabolic and signal transduction networks. In this paper, we presented an adapted Bayesian classifier based on tree augmented naïve Bayesian classifier to predict interface residues of protein-protein interaction sites. This classifier used fixed structure which could denote the correlation of sequence neighborhood and the character of protein's features. The conditional probability tables of the classifier were learned from the training dataset. By testing our approach on 81 hetero-complex chains, experimental results demonstrate the performance of our approach is indeed superior to current existing methods. The success of the predictions is validated by examining the predictions in the context of the three-dimensional structures of protein complexes. ©2009 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Inductive transfer learning for unlabeled target-domain via hybrid regularization

引用

Chinese Science Bulletin 2009年第11期54卷 2470-2478页

作者： ZHUANG FuZhen LUO Ping HE Qing SHI ZhongZhi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Hewlett Packard Labs China Beijing 100084 China Graduate University of Chinese Academy of Sciences Beijing 100190 China

Recent years have witnessed an increasing interest in transfer learning. This paper deals with the classification problem that the target-domain with a different distribution from the source-domain is totally unlabeled, and aims to build an inductive model for unseen data. Firstly, we analyze the problem of class ratio drift in the previous work of transductive transfer learning, and propose to use a normalization method to move towards the desired class ratio. Furthermore, we develop a hybrid regularization framework for inductive transfer learning. It considers three factors, including the distribution geometry of the target-domain by manifold regularization, the entropy value of prediction probability by entropy regularization, and the class prior by expectation regularization. This framework is used to adapt the inductive model learnt from the source-domain to the target-domain. Finally, the experiments on the real-world text data show the effectiveness of our inductive method of transfer learning. Meanwhile, it can handle unseen test points.

关键词：归纳学习正规化标签杂交归一化法预测概率文字资料归纳法

来源：评论

学校读者我要写书评

暂无评论

Face recognition using a hybrid model

Face recognition using a hybrid model

引用

38th Applied Imagery Pattern Recognition Workshop: Vision: Humans, Animals, and Machines, AIPRW 2009

作者： Wang, Yuheng Anderson, Peter G. Gaborski, Roger S. Golisano College of Computing and Information Sciences Rochester Institute of Technology Rochester NY 14623 United States Computer Science Department Rochester Institute of Technology Rochester NY 14623 United States Computer Science Department Laboratory of Intelligent Systems Rochester Institute of Technology Rochester NY 14623 United States

ISBN: (纸本)9781424451463

This paper introduces a hybrid face recognition model that combines biologically inspired features and Local Binary Features. The structure of the model is mainly based on the human visual ventral pathway. Previously, object-centered models focus on extracting global view-invariant representation of faces (I. Biederman, 1987) while feed-forward view-based models (HMAX model by Riesenhuber and Poggio, 1999) extract local features of faces by simulating responses of neurons in the human visual system. In this paper we first review the current main face recognition algorithms: Local Binary Pattern model and RandP model. This is followed by a detailed description of their implementation and advantages in overcoming intra-class variance. Results from our model are compared to the original Riesenhuber and Poggio model and Local Binary Pattern model (T. Ahonen et al, 2005). Then the paper will focus on our hybrid biological model which takes advantages of both structural information and biological features. Our model shows improved recognition rates and increased tolerance to intra-personal view differences.

关键词： Local binary pattern

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：