检索结果-内蒙古大学图书馆

Sequential em for unsupervised adaptive gaussian mixture model based classifier

6th international conference on machine learning and data mining in pattern recognition, MLDM 2009

作者： Awwad Shiekh Hasan, Bashar Gan, John Q. School of Computer Science and Electronic Engineering University of Essex Wivenhoe Park CO4 3SQ United Kingdom

ISBN: (纸本)3642030696

In this paper we present a sequential expectation maximization algorithm to adapt in an unsupervised manner a Gaussian mixture model for a classification problem. the goal is to adapt the Gaussian mixture model to cope with the non-stationarity in the data to classify and hence preserve the classification accuracy. Experimental results on synthetic data show that this method is able to learn the time-varying statistical features in data by adapting a Gaussian mixture model online. In order to control the adaptation method and to ensure the stability of the adapted model, we introduce an index to detect when the adaptation would fail. © 2009 Springer Berlin Heidelberg.

关键词： Maximum principle

来源：评论

学校读者我要写书评

暂无评论

Voted NER system using appropriate unlabeled data

Voted NER system using appropriate unlabeled data

引用

2009 Named Entities Workshop: Shared Task on Transliteration, NEWS 2009 at the Joint conference of the 47th Annual Meeting of the ACL and the 4th international Joint conference on Natural Language Processing of the AFNLP, ACL-IJCNLP 2009

作者： Ekbal, Asif Bandyopadhyay, Sivaji Dept. of Computer Science and Engg. Jadavpur University Kolkata700032 India

this paper reports a voted Named Entity recognition (NER) system with the use of appropriate unlabeled data. the proposed method is based on the classifiers such as Maximum Entropy (ME), Conditional Random Field (CRF) and Support Vector machine (SVM) and has been tested for Bengali. the system makes use of the language independent features in the form of different contextual and orthographic word level features along with the language dependent features extracted from the Part of Speech (POS) tagger and gazetteers. Context patterns generated from the unlabeled data using an active learning method have been used as the features in each of the classifiers. A semi-supervised method has been used to describe the measures to automatically select effective documents and sentences from unlabeled data. Finally, the models have been combined together into a final system by weighted voting technique. Experimental results show the effectiveness of the proposed approach with the overall Recall, Precision, and F-Score values of 93.81%, 92.18% and 92.98%, respectively. We have shown how the language dependent features can improve the system performance. © 2009 ACL and AFNLP.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Exploiting unsupervised and supervised classification for segmentation of the pathological lung in CT

引用

JOURNAL OF INSTRUMENTATION 2009年第7期4卷 P07013-P07013页

作者： Korfiatis, P. Kalogeropoulou, C. Daoussis, D. Petsas, T. Adonopoulos, A. Costaridou, L. Univ Patras Sch Med Dept Med Phys Patras 26500 Greece Univ Hosp Patras Dept Radiol Patras 26500 Greece Univ Hosp Patras Dept Rheumatol Patras 26500 Greece

Delineation of lung fields in presence of diffuse lung diseases (DLPDs), such as interstitial pneumonias (IP), challenges segmentation algorithms. To deal with IP patterns affecting the lung border an automated image texture classification scheme is proposed. the proposed segmentation scheme is based on supervised texture classification between lung tissue ( normal and abnormal) and surrounding tissue ( pleura and thoracic wall) in the lung border region. this region is coarsely defined around an initial estimate of lung border, provided by means of Markov Radom Field modeling and morphological operations. Subsequently, a support vector machine classifier was trained to distinguish between the above two classes of tissue, using textural feature of gray scale and wavelet domains. 17 patients diagnosed with IP, secondary to connective tissue diseases were examined. Segmentation performance in terms of overlap was 0.924 +/- 0.021, and for shape differentiation mean, rms and maximum distance were 1.663 +/- 0.816, 2.334 +/- 1.574 and 8.0515 +/- 6.549 mm, respectively. An accurate, automated scheme is proposed for segmenting abnormal lung fields in HRC affected by IP

关键词： data processing methods pattern recognition, cluster finding, calibration and fitting methods Computerized Tomography (CT) and Computed Radiography (CR)

来源：评论

学校读者我要写书评

暂无评论

machine learning framework for classification in medicine and biology

Machine learning framework for classification in medicine an...

引用

6th international conference on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems, CPAIOR 2009

作者： Lee, Eva K. Center for Operations Research in Medicine and HealthCare School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta GA 30332-0205 United States

ISBN: (纸本)3642019285

Systems modeling and quantitative analysis of large amounts of complex clinical and biological data may help to identify discriminatory patterns that can uncover health risks, detect early disease formation, monitor treatment and prognosis, and predict treatment outcome. In this talk, we describe a machine-learning framework for classification in medicine and biology. It consists of a pattern recognition module, a feature selection module, and a classification modeler and solver. the pattern recognition module involves automatic image analysis, genomic pattern recognition, and spectrum pattern extractions. the feature selection module consists of a combinatorial selection algorithm where discriminatory patterns are extracted from among a large set of pattern attributes. these modules are wrapped around the classification modeler and solver into a machine learning framework. the classification modeler and solver consist of novel optimization-based predictive models that maximize the correct classification while constraining the inter-group misclassifications. the classification/predictive models 1) have the ability to classify any number of distinct groups;2) allow incorporation of heterogeneous, and continuous/time-dependent types of attributes as input;3) utilize a high-dimensional data transformation that minimizes noise and errors in biological and clinical data;4) incorporate a reserved-judgement region that provides a safeguard against over-training;and 5) have successive multi-stage classification capability. Successful applications of our model to developing rules for gene silencing in cancer cells, predicting the immunity of vaccines, identifying the cognitive status of individuals, and predicting metabolite concentrations in humans will be discussed. We acknowledge our clinical/biological collaborators: Dr. Vertino (Winship Cancer Institute, Emory), Drs. Pulendran and Ahmed (Emory Vaccine Center), Dr. Levey (Neurodegenerative Disease and Alzheimer's Disease

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

thalamic bursts mediate pattern recognition

Thalamic bursts mediate pattern recognition

引用

international IEEE/EMBS conference on Neural Engineering, CNE

作者： Magnus Jandel Swedish Defence Research Agency Stockholm Sweden

A new functional model for burst firing in the dorsal thalamus is proposed where thalamocortical pattern recognition systems, based on kernel machine principles, are connected by burst signaling. the systems include input trapping in the dorsal thalamus, cortical learning state memory and processing in the thalamic reticular nucleus. Misclassified events are captured as training examples in the waking state and the pattern recognition systems are trained by extensive thalamic bursting in deep sleep.

关键词： pattern recognition Support vector machines Support vector machine classification Kernel Registers Neurons Digital relays Mathematical model Neural engineering machine learning

来源：评论

学校读者我要写书评

暂无评论

On Utilizing Optimal and Information theoretic Syntactic Modeling for Peptide Classification

On Utilizing Optimal and Information Theoretic Syntactic Mod...

引用

4th international conference pattern recognition in Bioinformatics

作者： Aygun, Eser Oommen, B. John Cataltepe, Zehra Istanbul Tech Univ Dept Comp Eng TR-80626 Istanbul Turkey Carleton Univ Sch Comp Sci Ottawa ON K1S 5B6 Canada Univ Agder Kristiansand Norway Istanbul Tech Univ Dept Comp Engn Istanbul Turkey

ISBN: (纸本)9783642040306

Syntactic methods in pattern recognition have been used extensively in bioinformatics, and in particular, in the analysis of gene and protein expressions, and in the recognition and classification of biosequences, these methods are almost universally distance-based. this paper concerns the use of an Optimal and Information theoretic (OIT) probabilistic model [11] to achieve peptide classification using the information residing in their syntactic representations. the latter has traditionally been achieved using the edit distances required in the respective peptide comparisons. We advocate that, one can model the differences between compared strings as a mutation model consisting of random Substitutions, Insertions and Deletions (SID) obeying the OIT model. thus, in this paper, we show that the probability measure obtained. from the OIT model can be perceived as a sequence similarity metric, using which a Support Vector machine (SVM)-based peptide classifier, referred to as OIT-SVM, can be devised. the classifier, which we have built has been tested for eight different "substitution" matrices and for two different data sets, namely, the HIV-1 Protease Cleavage sites and the T-cell Epitopes. the results show that the OIT model performs significantly better than the one which uses a Needleman-Wunsch sequence alignment score, and the peptide classification methods that previously experimented with the same two datasets.

关键词： Biological Sequence Analysis Optimal and Information theoretic Syntactic Classifcation Peptide Classification Sequence Processing Syntactic pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Static hand sign recognition using linear projection methods

Static hand sign recognition using linear projection methods

引用

international conference on Autonomous Robots and Agents, ICARA

作者： Nuwan Gamage Kuang Ye Chow Rini Akmeliawati Department of Electrical & Computer Systems Engineering Monash University Bandar Sunway Malaysia Department of Mechatronics Engineering International Islamic University Malaysia Gombak Malaysia

Shape matching is one of the more significant research topics in the fields of computer vision, pattern recognition and machine learning. Successful shape matching algorithms/ methods has a high potential for a wide variety of practical applications. In this paper, we present our effort on using linear projection methods for static hand sign recognition in Malaysian sign language. PCA and LPP methods have been used with a database of 240 hand shapes.

关键词： Shape Handicapped aids pattern matching Computer vision pattern recognition machine learning machine learning algorithms Application software Principal component analysis databases

来源：评论

学校读者我要写书评

暂无评论

Actas da 4a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2009

Actas da 4a Conferencia Iberica de Sistemas e Tecnologias de...

引用

4a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2009 - 4th Iberian conference on Information Systems and Technologies, CISTI 2009

ISBN: (纸本)9789899624719

the proceedings contain 149 papers. the topics discussed include: rule based creative pattern generation for visual composition;date pre-processing issues: a case study for database marketing;SWDSS - hints to transform single system into a software product line;citizen participation in city planning and public decision assisted with ontologies and 3D semantics;classification of facial expressions using data mining and machine learning algorithms;applying hidden Markov models to process mining;on the strengthening of OPENID authentication mechanisms with the Portuguese citizen card;high quality video streaming over 3G+ mobile networks;a proposed unified communications platform based on open source technologies;and contribution to experimental performance evaluation of point-to-point links using WIMAX and Wi-Fi technologies.

关键词：

来源：评论

学校读者我要写书评

暂无评论

learning with a quadruped chopstick robot

引用

6th international conference on machine learning and data mining in pattern recognition, MLDM 2009

作者： Lee, Wei-Chung Chen, Jong-Chen Wu, Shou-Zhe Lin, Kuo-Ming National Yunlin University of Science and Technology Taiwan

ISBN: (纸本)3642030696

Organisms exhibit a close structure-function relationship and a slight change in structure may in turn change their outputs accordingly [1]. this feature is important as it is the main reason why organisms have better malleability than computers in dealing with environmental changes. A quadruped chopstick robot controlled by a biologically-motivated neuromolecular model, named Miky, has been developed. Miky's skeleton and its four feet were comprised of 16 deposable chopsticks, with each foot being controlled by an actuator (motor). the neuromolecular model is a multilevel neural network which captures the biological structure-function relationship and serves to transform signals sent from its sensors into a sequence of signals in space and time for controlling Miky's feet (through actuators). the task is to teach Miky to walk, jump, pace, gallop, or make a turn. Our experimental result shows that Miky exhibits a close structure-function relationship that allows it to learn to accomplish these tasks in a continuous manner. © 2009 Springer Berlin Heidelberg.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

Automatic Categorization of Tags in Collaborative Environments

Automatic Categorization of Tags in Collaborative Environmen...

引用

4th international conference of Collaborative Computing: Networking, Applications and Worksharing

作者： Wang, Qihua Jin, Hongxia Nusser, Stefan Purdue Univ W Lafayette IN 47907 USA IBM Almaden Res Ctr San Jose CA USA

ISBN: (纸本)9783642033537

Tagging allows individuals to use whatever terms they think are appropriate to describe an item. With the growing popularity of tagging, more and more tags have been collected by a variety of applications. An item may be associated with tags describing its different aspects, such as appearance, functionality, and location. However, little attention has been paid in the organization of tags;in most tagging systems, all the tags associated with an item are listed together regardless of their meanings. When the number of tags becomes large, finding useful information with regards to a certain aspect of an item becomes difficult. Improving the organization of tags in existing tagging systems is thus highly desired. In this paper, we propose a hierarchical approach to organize tags. In our approach, tags are placed into different categories based on their meanings. To find information with respect to a certain aspect of an item, one just needs to refer to its associated tags in the corresponding category. Since existing applications have already collected a large number of tags, manually categorizing all the tags is infeasible. We propose to use data-mining and machine-learning techniques to automatically and rapidly classify tags in tagging systems. A prototype of our approaches has been developed for a real-word tagging system.

关键词： Tagging Web 2.0 Social Network Categorization machine learning

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：