检索结果-内蒙古大学图书馆

International Conference on Computational Intelligence and Security

作者： Chunyan Liu Conghui Zhu Tiejun Zhao Dequan Zheng MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

Online social media has become one of the most important ways people communicate, while how to find valuable information from huge amounts of data becomes a key problem. We present a novel topic extraction method that employs topic value of each words and social model attributes as additional features based on the multi-document summarization. The experimental results show that the multi-document summarization with the topic and the sociality are helpful to extract topics from social media.

关键词： Blogs Social network services Feature extraction Dictionaries Data mining Media Natural language processing

来源：评论

学校读者我要写书评

暂无评论

A tentative study on rule based metaphor identification

A tentative study on rule based metaphor identification

引用

International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)

作者： Yu Zhang Muyun Yang Xi Chen Sheng Li Tiejun Zhao MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China School of Foreign Languages Harbin Institute of Technology Harbin China

As one of the challenging issues in the field of Natural language processing (NLP), metaphor has aroused substantial attention among researchers in recent years. Many models and methods have been proposed for proper understanding of metaphors. But the automatic identification of metaphor is less touched. This paper presents a tentative study on the metaphor identification based on rules, and the results on a small scale corpus are provided.

关键词： Semantics Natural language processing Computational linguistics Tin Artificial intelligence Silicon Context

来源：评论

学校读者我要写书评

暂无评论

Forced decoding for minimum error rate training in statistical machine translation

引用

Journal of Computational Information Systems 2012年第2期8卷 861-868页

作者： Liang, Huashen Zhang, Min Zhao, Tiejun MOE-MS Key Lab. of Natural Language Processing and Speech Harbin Institute of Technology Harbin 150001 China Institute for Infocomm Research FusionopolisWay #21-01 Connexis Singapore 138632 Singapore

We present a forced decoding approach for the tuning process in statistical machine translation. Unlike the traditional discriminative approaches, the forced decoding system can take advantage of the reference of development set to generate n-best list in the decoding stage, and then merge this list with original n-best list generated by decoder in the tuning stage. We find that traditional minimum error rate training process can not find good translations for long sentences. While the forced decoding system knows the references, it could generate translations that are more similar to the references, which means the results would have higher BLEU score. As a supplement, the result of forced decoding system would lead the tuning process to a better point. 1553-9105/Copyright © 2012 Binary Information Press.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Syntax encapsulated phrase model for statistical machine translation

Syntax encapsulated phrase model for statistical machine tra...

引用

International Conference on Information Science and technology (ICIST)

作者： Huashen Liang Tiejun Zhao Yongzeng Xue MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China Department of New Media and Art Harbin Institute of Technology Harbin China

In the past few years, much attention has been paid on extending phrase-based statistical machine translation with syntactic structures. In this paper we introduce a novel syntax encapsulated phrase(SEP) model, in which treebank tag sequences are employed to decorate the bilingual phrase pairs. We use tag sequences, instead of phrase pairs, to train the lexicalized reordering model. Since the number of treebank tags is much smaller than the number of words, the tag sequence based reordering model is smaller and more accurate than the phrase based reordering model. Experiments were carried out on four types of models: the phrase model, the hierarchical phrase model, the POS tag encapsulated phrase(PTEP) model and the syntactic tag encapsulated phrase(STEP) model. The STEP model obtained higher BLEU-4 score than other models on NIST 2005 MT task.

关键词： Syntactics Training Vectors Computational modeling NIST Data models Mathematical model

来源：评论

学校读者我要写书评

暂无评论

An effective approach to pedestrian detection in thermal imagery

An effective approach to pedestrian detection in thermal ima...

引用

International Conference on Natural Computation (ICNC)

作者： Wei Li Dequan Zheng Tiejun Zhao Mengda Yang MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China Institute for Interdisciplinary Information Science Tsinghua University Beijing China

In this paper, an integrated algorithm to detect humans in thermal imagery was introduced. In recent years, histogram of oriented gradient (HOG) is a quite popular algorithm for person detection in visible imagery. We implement the pedestrian detection in infrared imagery with this algorithm by adjusting the parameters. Simultaneously, we have increased some other geometric characteristics, such as mean contrast, which is used as features for the detection. After analyzing the property of the infrared imagery, which is designed to meet the shortfall of the HOG in infrared imagery, the combined vectors are fed to a linear SVM for object/non-object classification and we get the detector at the same time. After that, the detection window is scanned across the image at multiple positions and scales, which is followed by the combination of the overlapping detections. At last, a pedestrian is described by a final detection, and we have detected the pedestrians in the thermal imagery. Experimental results with OSU Thermal Pedestrian Database are reported to demonstrate the excellent performance of our algorithms.

关键词： Feature extraction Histograms Detectors Training Support vector machines Computer vision Sensitivity

来源：评论

学校读者我要写书评

暂无评论

Event recognition based on time series characteristics

Event recognition based on time series characteristics

引用

2011 8th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2011, Jointly with the 2011 7th International Conference on Natural Computation, ICNC'11

作者： Li, Fenghuan Zheng, Dequan Zhao, Tiejun MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

ISBN: (纸本)9781612841816

Event recognition and temporal information analysis are important subtasks in information extraction (IE). In this paper, event recognition based on time series characteristics is proposed. In the pipeline of event recognition, trigger word table is extracted from training corpus and extended based on the field and thesaurus, which is regarded as a priori knowledge. Then event recognition is carried out using trigger words and support vector machine (SVM). Temporal expressions are normalized primarily when recognizing event time. Especially, keywords on time and their priorities are taken into account. Finally, events are sorted by time series characteristics. The results show that methods proposed in this paper are valid and effective. © 2011 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Search results optimization method combined with multi-features

Search results optimization method combined with multi-featu...

引用

2011 8th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2011, Jointly with the 2011 7th International Conference on Natural Computation, ICNC'11

作者： Qin, Yanxia Zheng, Dequan Xu, Bing MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

ISBN: (纸本)9781612841816

The optimization of search results has always been the research hotspot in the area of search engine. More concretely, topic partition by clustering proved to be a good way. However, the clusters, some of which still contain a lot of documents, implicitly limit the users' retrieval speed. Meanwhile we find that the information of documents' features have good effects on the document ranking. To address the issue, we try to apply the multi-features to search results after the process of clustering. Statistic and semantic information of the multi-features are fully used to re-rank the documents. Related experiments show that our approach outperforms that of single clustering much. The evaluation indicators' rising shows that the Top N results satisfy the users' need more. © 2011 IEEE.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Study on Chinese person name disambiguation based on multi-stage strategy

Study on Chinese person name disambiguation based on multi-s...

引用

2011 8th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2011, Jointly with the 2011 7th International Conference on Natural Computation, ICNC'11

作者： Han, Wei Xu, Bing Zhao, Tiejun MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

ISBN: (纸本)9781612841816

Namesake is a very common phenomenon both in real world and in the Internet. This paper combines the problem of name disambiguation with clustering technique and attempts to achieve the purpose of person name disambiguation through clustering technique by putting different texts pointing to the same person to one cluster. In this paper, we propose a multi-stage strategy for text clustering. In the first stage, the bias classifier is used to generate the initial categories. The second stage is a re-clustering process of agglomerative hierarchical clustering to achieve the name disambiguation. Experimental results show that the method proposed in this paper can effectively improve the accuracy of person name disambiguation. © 2011 IEEE.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised learning for word sense disambiguation using parallel corpora

Semi-supervised learning for word sense disambiguation using...

引用

2011 8th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2011, Jointly with the 2011 7th International Conference on Natural Computation, ICNC'11

作者： Yu, Mo Wang, Shu Zhu, Conghui Zhao, Tiejun MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

ISBN: (纸本)9781612841816

The Application of word sense disambiguation (WSD) methods based on supervised machine learning are limited by the difficulties in defining sense tags and acquiring labeled data for training. In this paper, the two problems of WSD are solved in a semi-supervised learning framework with the help of parallel corpora. The sense tags are defined automatically according to the results of word alignment on the parallel corpora. And label propagation, a graph-based semi-supervised algorithm, is employed. The experiments show that our method achieves great improvement on Chinese WSD tasks and the performances get significant growth when the scale of monolingual sentences is increasing. © 2011 IEEE.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Research on machine learning method-based combination forecasting model and its application

Research on machine learning method-based combination foreca...

引用

2011 8th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2011, Jointly with the 2011 7th International Conference on Natural Computation, ICNC'11

作者： Sun, Zhenlong Zhu, Conghui Xu, Bing Li, Sheng MOE-MS Key Laboratory of Natural Language Processing and Speech Harbin Institute of Technology Harbin China

ISBN: (纸本)9781612841816

A novel combination forecasting model is presented in this paper, which combines single ones based on machine learning. The model has been applied to the prediction of five cities' election in Taiwan with combining the exposure rate and the approval rate, which obtains good results. The exposure rate is the frequency of a candidate's appearances in the news and approval rate is the proportion of the positive information of a candidate. And the polarity of a review is predicted by sentiment classification based on machine learning techniques. A novel method of feature extraction is used in sentiment classification, which makes the classifier effectively assign the review a type of polarity. Meanwhile, this paper proposes a method of feature clustering and extending based on the synonym dictionary, which obviously reduces the dimension of feature vector and improve the F-score of sentiment classification. © 2011 IEEE.

关键词： Feature extraction

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：