检索结果-内蒙古大学图书馆

Active learning for imbalanced sentiment classification

Active learning for imbalanced sentiment classification

2012 Joint Conference on Empirical Methods in Natural Language processing and Computational Natural Language Learning, EMNLP-CoNLL 2012

作者： Li, Shoushan Ju, Shengfeng Zhou, Guodong Li, Xiaojun Natural Language Processing Lab. School of Computer Science and Technology Soochow University Suzhou 215006 China College of Computer and Information Engineering Zhejiang Gongshang University Hangzhou 310035 China

ISBN: (纸本)9781937284435

Active learning is a promising way for sentiment classification to reduce the annotation cost. In this paper, we focus on the imbalanced class distribution scenario for sentiment classification, wherein the number of positive samples is quite different from that of negative samples. This scenario posits new challenges to active learning. To address these challenges, we propose a novel active learning approach, named co-selecting, by taking both the imbalanced class distribution issue and uncertainty into account. Specifically, our co-selecting approach employs two feature subspace classifiers to collectively select most informative minority-class samples for manual annotation by leveraging a certainty measurement and an uncertainty measurement, and in the meanwhile, automatically lab.l most informative majority-class samples, to reduce human-annotation efforts. Extensive experiments across four domains demonstrate great potential and effectiveness of our proposed co-selecting approach to active learning for imbalanced sentiment classification. © 2012 Association for Computational Linguistics.

关键词： Uncertainty analysis

来源：评论

学校读者我要写书评

暂无评论

Sparse sampling based clutter spectrum compensation method for bistatic airborne radar

引用

Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics 2012年第6期34卷 1120-1125页

作者： Liu, Jin-Hui Liao, Gui-Sheng Li, Ming Institute of Computer Peripherals Xidian University Xi'an 710071 China National Lab. of Radar Signal Processing Xidian University Xi'an 710071 China

In a bistatic airborne radar, since the range-dependence (RD) problem exists in the received ground clutter spectrum, the clutter covariance matrix, which is estimated from the average of the clutter data of each range cell, cannot be accurately estimated, and it makes the clutter suppression performance of space-time adaptive processing (STAP) decline. The registration based compensation method compensates the RD problem by sampling along the direction-Doppler curve of the clutter data, but requires heavy computation load due to the large sampling points. Aiming at this problem, the sparse sampling clutter spectrum compensation method is proposed by using the sparse characteristic of the clutter complex amplitude, which is modulated by the transmitter and receiver antennas in the bistatic airborne radar. The proposed method reduces the sampling point number and decreases the computation load of the clutter data reconstruction by setting the sampling threshold. The computer simulation results show that the proposed method can reduce the computation load, mitigate the RD problem of the ground clutter in bistatic airborne radar and thus improve the clutter rejection performance of STAP.

关键词： Space time adaptive processing

来源：评论

学校读者我要写书评

暂无评论

Automatic speaker recognition for mobile communications using AMR-WB speech coding

Automatic speaker recognition for mobile communications usin...

引用

International Symposium on signal processing and Its Applications (ISSPA)

作者： Meriem Fedila Abderrahmane Amrouche Speech Communication & Signal Processing Lab. Faculty of Electronics and Computer Science Algiers Algeria

ISBN: (纸本)9781467303811

The aim of this paper is to investigate the influence of Adaptative Multi-Rate Wideband (AMR-WB) speech coding on Distributed Speaker Recognition (DSR). The main goal is to improve speaker recognition performance without resynthesizing the speech waveform in mobile communications. For this purpose, we have implemented a method in order to extract the acoustic features in the compressed domain, directly from encoded bitstream. The obtained results, using ARADIGIT database, show that the proposed approach, with GMM and SVM models, using ISF (Immittance Spectral Frequency) parameters through a noised channel (AWGN and Rayleigh) are promising.

关键词： Speech Speaker recognition Speech coding Feature extraction Support vector machines Databases Vectors

来源：评论

学校读者我要写书评

暂无评论

Two-microphone speech enhancement using a learned binary mask

Two-microphone speech enhancement using a learned binary mas...

引用

20th Iranian Conference on Electrical Engineering, ICEE 2012

作者： Abdipour, Roohollah Akbari, Ahmad Rahmani, Mohsen Nasersharif, Babak Audio and Speech Processing Lab. School of Computer Engineering Iran University of Science and Technology Iran Computer Engineering Department Faculty of Engineering Arak University Iran Electrical and Computer Engineering Department K.N.Toosi University of Technology Tehran Iran

ISBN: (纸本)9781467311489

Ideal binary mask speech enhancement is shown to increase the speech quality as well as speech intelligibility. But, this property depends highly on the accurate separation of speech and masker time-frequency units of the input spectrum, which is a difficult task in real situations. Ordinary binary mask methods are single-microphone methods and so, can obtain little information from the environment. In this paper, we devise a two-microphone method that uses a classifier to distinguish speech-dominated and masker-dominated time-frequency units. The classifier uses simply computable two-microphone features which enable it to be used in real-time scenarios. These proposed features empower the classifier to reach to classification accuracies near 80%. This high accuracy in turn, empowers the Ideal binary mask method to obtain higher SNRI and NPLR values in comparison to state-of-the-art noise reduction methods. These results indicate that the proposed two-microphone features have high information content for speech/masker separation. © 2012 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Improved region local binary patterns for image retrieval

Improved region local binary patterns for image retrieval

引用

computer Science and Information Engineering, CSIE 2012

作者： Wu, Xiaosheng Sun, Junding School of Computer Science and Technology Henan Polytechnic University Jiaozuo 454003 China Image Processing and Image Communication Lab. Nanjing University of Post and Telecommunication Nanjing 210003 China

ISBN: (纸本)9783642301254

Two extended patterns, direction local binary pattern (D-LBP) and variance direction local binary pattern (vD-LBP) were introduced in the paper. More texture features are obtained in the new operators by introducing a variant threshold. The descriptors mentioned in the paper were firstly tested and evaluated on CUReT texture image database. Experimental results show that the two extended operators give better performance than D-LBP and vD-LBP respectively. Finally, the operators were used for trace fossils image retrieval and the results denote that the proposed operator, tvD-LBP is the most effective method for trace fossils image description in the four descriptors. © 2012 Springer-Verlag GmbH.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

Energy-efficient network routing with discrete cost functions

Energy-efficient network routing with discrete cost function...

引用

9th Annual Conference on Theory and Applications of Models of Computation, TAMC 2012

作者： Wang, Lin Fernández Anta, Antonio Zhang, Fa Hou, Chenying Liu, Zhiyong Key Lab. of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Institute IMDEA Networks Spain China State Key Lab. for Computer Architecture Institute of Computing Technology Chinese Academy of Sciences China Graduate University of Chinese Academy of Sciences China

ISBN: (纸本)9783642299513

Energy consumption is an important issue in the design and use of networks. In this paper, we explore energy savings in networks via a rate adaptation model. This model can be represented by a cost-minimization network routing problem with discrete cost functions. We formulate this problem as an integer program, which is proved to be NP-hard. Then a constant approximation algorithm is developed. In our proposed method, we first transform the program into a continuous-cost network routing problem, and then we approximate the optimal solution by a two-step rounding process. We show by analysis that, for uniform demands, our method provides a constant approximation for the uniform network routing problem with discrete costs. A bicriteria network routing problem is also developed so that a trade-off can be made between energy consumption and network delay. Analytical results for this latter model are also presented. © 2012 Springer-Verlag.

关键词： Network routing

来源：评论

学校读者我要写书评

暂无评论

A novel EMD-based Common Spatial Pattern for motor imagery brain-computer interface

A novel EMD-based Common Spatial Pattern for motor imagery b...

引用

IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering

作者： He, Wei Wei, Pengfei Wang, Liping Zou, Yuexian Shenzhen Key Lab. of Neuropsychiatric Modulation Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen 518055 China Advanced Digital Signal Processing Laboratory Peking University Shenzhen Graduate School Shenzhen 518055 China

ISBN: (纸本)9781457721779

Common Spatial Pattern (CSP) algorithm is a commonly used effective feature extraction method in motor imagery (MI) electroencephalogram (EEG) based brain computer interface (BCI). The motor imagery patterns extracted by CSP are associated with variations in subject-specific frequency bands power. Therefore, optimizing frequency band carrying MI intention is required by the CSP method. However, the frequency band is usually divided manually and evaluated which does make use of the EEG signal property and also with low efficiency. In this paper, we propose a novel Empirical Mode Decomposition (EMD) based CSP method to realize the data-related and adaptive frequency band selection. Specifically, the intrinsic mode functions (IMFs) decomposed from the EMD and the amplitude modulated signal by instantaneous amplitude (IA) calculated from Hilbert Transform have been fully explored and employed. Moreover, the intensive experiments have been conducted to evaluate the proposed method. From the experiment results, we observed that the EMD based CSP method enhances the classification accuracy in BCI competition IV dataset I for all subjects and a paired t-test shows a significant difference level. © 2012 IEEE.

关键词： Empirical mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Identifying high-impact sub-structures for convolution kernels in document-level sentiment classification

Identifying high-impact sub-structures for convolution kerne...

引用

50th Annual Meeting of the Association for Computational Linguistics, ACL 2012

作者： Tu, Zhaopeng He, Yifan Foster, Jennifer Van Genabith, Josef Liu, Qun Lin, Shouxun Key Lab. of Intelligent Info. Processing Institute of Computing Technology CAS China Computer Science Department New York University United States School of Computing Dublin City University Ireland

ISBN: (纸本)9781937284251

Convolution kernels support the modeling of complex syntactic information in machinelearning tasks. However, such models are highly sensitive to the type and size of syntactic structure used. It is therefore an important challenge to automatically identify high impact sub-structures relevant to a given task. In this paper we present a systematic study investigating (combinations of) sequence and convolution kernels using different types of substructures in document-level sentiment classification. We show that minimal sub-structures extracted from constituency and dependency trees guided by a polarity lexicon show 1.45 point absolute improvement in accuracy over a bag-of-words classifier on a widely used sentiment corpus. © 2012 Association for Computational Linguistics.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

The structure of the sparse basis function in compressive sensing for fast analysis of wide-angle monostatic scattering problems

The structure of the sparse basis function in compressive se...

引用

2012 International Conference on Microwave and Millimeter Wave technology, ICMMT 2012

作者： Du, Hong-Mei Chen, Ming-Sheng Wu, Xian-Liang Key Lab. of Intelligent Computing and Signal Processing Anhui University Ministry of Education Hefei 230029 China Department of Physics and Electronic Engineering Hefei Normal University Hefei 230601 China University of Science and Technology of China Hefei 230026 China

ISBN: (纸本)9781467321839

In the theory of compressive sensing, the selection of the basis functions directly affects the sparse transformation, observation number and reconstruction accuracy. In this paper, we introduce the structure of three sparse basic functions, which are used to analyze the electromagnetic scattering problems over a wide incident angle by the method of moment and compressive sensing. According to the characteristics of the induced current of wide angle, the best basic function is selected. Compared with the DCT, the accuracy of the reconstruction for different shape object are improved and the observation number reduces about 50%. © 2012 IEEE.

关键词： Method of moments

来源：评论

学校读者我要写书评

暂无评论

Using simple relays to improve physical-layer security

Using simple relays to improve physical-layer security

引用

2012 1st IEEE International Conference on Communications in China, ICCC 2012

作者： Zheng, Gan Li, Jiangyuan Wong, Kai-Kit Petropulu, Athina P. Ottersten, Björn University of Luxembourg Luxembourg Luxembourg Department of Electrical and Computer Engineering Rutgers-The State University of New Jersey New Brunswick NJ United States Department of Electronic and Electrical Engineering University College London London United Kingdom Signal Processing Lab. ACCESS Linnaeus Center KTH Royal Institute of Technology Sweden

ISBN: (纸本)9781467328159

This paper studies different uses of two cooperating relays to improve the secrecy rate of a wiretap channel. These two relays are assumed to perform only simple functions: either amplify-and-forward (AF) or jamming. Complex functions such as decode-and-forward (DF) are not considered. We study three modes of cooperation: i) cooperative jamming (CJ), ii) AF-aided beamforming and iii) mixed AF-aided beamforming and CJ, all with individual relay power constraints. While i) is known in the literature, our efforts are spent on ii) and iii). In particular, for iii), we assume that the jamming signals in two communication stages are correlated, giving rise to improved performances. We also propose a heuristic approach for selecting the appropriate cooperating mode. Simulation results illustrate the performance gain of each scheme under different channel conditions and the effectiveness of the proposed mode selection method. © 2012 IEEE.

关键词： Beamforming

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：