检索结果-内蒙古大学图书馆

INFORMATION RETRIEVAL FOR SHORT DOCUMENTS

Journal of Electronics(China) 2006年第6期23卷 933-936页

作者： Qi Haoliang Li Mu Gao Jianfeng Li Sheng Ministry of Education - Microsoft Key Laboratory of Natural Language Processing and Speech （Harbin Institute of Technology Harbin 150001 China） Mtcrosoft Research Asia Beijing 100080 China Microsoft Research Redmond WA 98052 USA

The major problem of the most current approaches of information models lies in that individual words provide unreliable evidence about the content of the texts. When the document is short, e.g. only the abstract is available, the word-use variability problem will have substantial impact on the Information Retrieval （IR） performance. To solve the problem, a new technology to short document retrieval named Reference Document Model （RDM） is put forward in this letter. RDM gets the statistical semantic of the query/document by pseudo feedback both for the query and document from reference documents. The contributions of this model are three-fold：（1） Pseudo feedback both for the query and the document; （2） Building the query model and the document model from reference documents; （3） Flexible indexing units, which can be ally linguistic elements such as documents, paragraphs, sentences, n-grams, term or character. For short document retrieval, RDM achieves significant improvements over the classical probabilistic models on the task of ad hoc retrieval on Text REtrieval Conference （TREC） test sets. Results also show that the shorter the document, the better the RDM performance.

关键词： Information retrieval Short documents Reference Document Model （RDM）

来源：评论

学校读者我要写书评

暂无评论

Two-stage approach to full Chinese parsing

引用

High technology Letters 2005年第4期11卷 359-363页

作者：曹海龙 Zhao Tiejun Yang Muyun Li Sheng MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 P.R. China

natural language parsing is a task of great importance and extreme difficulty. In this paper, we present a full Chinese parsing system based on a two-stage approach. Rather than identifying all phrases by a uniform model, we utilize a divide and conquer strategy. We propose an effective and fast method based on Markov model to identify the base phrases. Then we make the first attempt to extend one of the best English parsing models i.e. the head-driven model to recognize Chinese complex phrases. Our two-stage approach is superior to the uniform approach in two aspects. First, it creates synergy between the Markov model and the head-driven model. Second, it reduces the complexity of full Chinese parsing and makes the parsing system space and time efficient. We evaluate our approach in PARSEVAL measures on the open test set, the parsing system performances at 87.53% precision, 87.95% recall.

关键词： natural language processing systems parsing markov model pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Chinese POS tagging based on bilexical co-occurrences

Chinese POS tagging based on bilexical co-occurrences

引用

International Conference on Machine Learning and Cybernetics, ICMLC 2005

作者： Cao, Hai-Long Zhao, Tie-Jun Sheng, Li Sun, Jun Zhang, Chun-Xiang MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 China

ISBN: (纸本)078039092X

Chinese part of speech tagging is the basis of Chinese information processing. This paper proposes a method based on bilexical co-occurrences to tag Chinese text. The standard hidden Markov model assumes the transition between states (part of speech) is independent of the observation (word) sequence and the generation of a new observation is independent of other observations. In fact, Chinese text does not satisfy this assumption. Based on hidden Markov model, the effect of the words in the context on the decision of part of speech is also considered. The discriminative ability of the model is improved. Deleted interpolation is utilized to mitigate the data sparseness problem. We evaluate the proposed model on PFR China Daily corpus. The tagging accuracy is 99.09% on closed test and 96.37% on open test. ©2005 IEEE.

关键词： speech synthesis

来源：评论

学校读者我要写书评

暂无评论

Research on an ontology-based language model

Research on an ontology-based language model

引用

International Conference on Chinese Computing 2005, ICCC 2005

作者： Zheng, Dequan Yu, Hao Zhao, Tiejun Li, Sheng Peng, Yuan MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Box 321 Harbin150001 China

ISBN: (纸本)9810530072

Ontology is widely applied to various areas of natural language processing, which regarded as a modeling tool of conceptual model of describing information in the level of semantic and knowledge. In this paper, we put forward an ontology-based language model with the ontological character, which is language-independent and can be used in cross-language information processing. Construct an ontology representation of feature information using a similarity framework;get the usage of feature information in using semantics, pragmatics, syntactic, etc. by specifically machine learning method and express as abstract ontology knowledge of feature information. processing actual document, the usage of feature information will be got in actual document and the similarity can be evaluated with ontology knowledge. We did three experiments in Chinese, i.e. IR, text similarity computing, specific information identification. The preliminary experiment testifies that recall and precision is advanced.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

High precision English base noun phrase identification based on "waterfall" model

High precision English base noun phrase identification based...

引用

International Conference on Machine Learning and Cybernetics, ICMLC 2005

作者： Liang, Ying-Hong Zhao, Tie-Jun Yu, Hao Yao, Jian-Min MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 School of Information and Computer Engineering North East Forestry University Harbin 150040

ISBN: (纸本)078039092X

Based on classical model that used by software exploitation in the subject of Software Engineering - waterfall model, a high precision model for English noun phrase identification is presented. In this model, three important features (interior structure, context information and boundary character) in base noun phrase identification are orderly used, and the relative method (the rule method, the transfer- based and error-driven method, and the boundary statistic method) are used from top to bottom. Thus the precision of base noun phrase identification improved steadily. The "waterfall" model combined the rule method and the statistical method so as to make them complement each other. The precision of base noun phrase identification achieves 98.10% and the F score is 95.25%. Compared to other method, our method achieves the highest precision and F score. © 2005 IEEE.

关键词： Information analysis

来源：评论

学校读者我要写书评

暂无评论

Distributed english text chunking using multi-agent based architecture

Distributed english text chunking using multi-agent based ar...

引用

4th Mexican International Conference on Artificial Intelligence, MICAI 2005

作者： Liang, Ying-Hong Zhao, Tie-Jun MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 School of Information and Computer Engineering North East Forestry University Harbin 150080

ISBN: (纸本)3540298967

The traditional English text chunking approach identifies phrases by using only one model and phrases with the same types of features. It has been shown that the limitations of using only one model are that: the use of the same types of features is not suitable for all phrases, and data sparseness may also result. In this paper, the Distributed Multi-Agent based architecture approach is proposed and applied in the identification of English phrases. This strategy put phrases into agents according to their sensitive features and identifies different phrases in parallel, where the main features are: one, easy and quick communication between phrases;two, avoidance of data sparseness. By applying and testing the approach on the public training and test corpus, the F score for arbitrary phrases identification using Distributed Multi-Agent strategy achieves 95.70% compared to the previous best F score of 94.17%. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

A multi-agent strategy for Chinese text chunking

A multi-agent strategy for Chinese text chunking

引用

International Conference on Machine Learning and Cybernetics, ICMLC 2005

作者： Liang, Ying-Hong Zhao, Tie-Jun Mao, Lei MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 School of Information and Computer Engineering North East Forestry University Harbin 150040

ISBN: (纸本)078039092X

Traditional Chinese text chunking approach is to identify phrases using only one model and same features. It is shown that one model couldn't comprise each phrase's characteristics, and same features are not suitable to all phrases, data sparseness also appears. Multi-agent strategy uses several model and sensitive features of each phrase to identify different phrases. This paper describes the Multi-agent strategy applied in the identification of Chinese phrases whose main features are: 1) easy and quick communication between phrases;2) avoidance of data sparseness. Through testing on Chinese Penn Treebank, F score of Chinese text chunking using Multi-agent strategy achieves to 95.82%, which is higher than the best result that has been reported. © 2005 IEEE.

关键词： Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Machine learning for automatic acquisition of Chinese linguistic ontology knowledge

Machine learning for automatic acquisition of Chinese lingui...

引用

International Conference on Machine Learning and Cybernetics, ICMLC 2005

作者： Zheng, De-Quan Zhao, Tie-Jun Yu, Feng Sheng-Li School of Computer Science and Engineering Harbin University of Commerce Harbin 150076 China MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 China

ISBN: (纸本)078039092X

Due to the complexity and flexibility of natural language, automatic linguistic knowledge acquisition and its application research becomes difficult. In this paper, we present a machine learning method to automatically acquire Chinese linguistic ontology knowledge from typical corpus. This study, first, defined the description frame of Chinese linguistic ontology knowledge, and then, automatically acquired the usage of a Chinese word with its co-occurrence of context in using semantic, pragmatics, syntactic, etc from the corpus, final, the above information and their representation will acted as Chinese linguistic ontology knowledge bank. We completed two groups of experiments, i.e. documents similarity computing, text reordering for information retrieval. Compared with previous works, the proposed method solves the inferior Precision of nature language processing. ©2005 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Research on automatic acquisition of translation template based on error-driven learning method

Research on automatic acquisition of translation template ba...

引用

International Conference on Machine Learning and Cybernetics, ICMLC 2005

作者： Zhang, Chun-Xiang Li, Sheng Zhao, Tie-Jun Cao, Hai-Long MOE-MS Key Laboratory of Natural Language Processing and Speech School of Computer Science and Technology Harbin Institute of Technology 150001 China School of Computer Science and Technology Harbin Institute of Technology 150001 China

ISBN: (纸本)078039092X

Automatic acquisition of translation templates is important for MT system to improve its translation quality and its ability of adaptation to new domain. In this paper, translation equivalences are obtained from translation corresponding trees of bilingual sentence pairs. Error-driven learning method is employed to acquire templates from extracted equivalences. At the same time, optimization method based on automatic translation evaluation is used to clean these templates. Then they are applied to a transfer-based MT system, and "863" dialog corpus in 2003 is used for open test. Experimental results show that the performance of new acquired templates exceeds that of original ones. Combination of new acquired templates and original ones makes 5-gram Nist assessment score of open test corpus improve by 8.11%. ©2005 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

A multi-agent strategy for Chinese text chunking

A multi-agent strategy for Chinese text chunking

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Ying-Hong Liang Tie-Jun Zhao Lei Mao MOE-MS Key Laboratory of Natural Language Processing and Speech in Harbin Institute of Technology Harbin China MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

Traditional Chinese text chunking approach is to identify phrases using only one model and same features. It is shown that one model couldn't comprise each phrase's characteristics, and same features are not suitable to all phrases, data sparseness also appears. Multi-agent strategy uses several model and sensitive features of each phrase to identify different phrases. This paper describes the multi-agent strategy applied in the identification of Chinese phrases whose main features are: 1) easy and quick communication between phrases; 2) avoidance of data sparseness. Through testing on Chinese Penn Treebank, F score of Chinese text chunking using multi-agent strategy achieves to 95.82%, which is higher than the best result that has been reported.

关键词： natural language processing Data mining Text processing Laboratories speech processing Forestry Electronic mail Testing Information analysis Statistical analysis

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：