检索结果-内蒙古大学图书馆

20th International Conference on Pattern Recognition, ICPR 2010

作者： Dimitrovski, Ivica Kocev, Dragi Loskovska, Suzana Džeroski, Sašo Department of Knowledge Technologies Jožef Stefan Institute Jamova cesta 39 1000 Ljubljana Slovenia Department of Computer Science Faculty of Electrical Engineering and Information Technology Karpoš bb 1000 Skopje Macedonia

ISBN: (纸本)3642177107

In this paper, we present a hierarchical multi-label classification system for visual concepts detection and image annotation. Hierarchical multi-label classification (HMLC) is a variant of classification where an instance may belong to multiple classes at the same time and these classes/labels are organized in a hierarchy. The system is composed of two parts: feature extraction and classification/annotation. The feature extraction part provides global and local descriptions of the images. These descriptions are then used to learn a classifier and to annotate an image with the corresponding concepts. To this end, we use predictive clustering trees (PCTs), which are able to classify target concepts that are organized in a hierarchy. Our approach to HMLC exploits the annotation hierarchy by building a single predictive clustering tree that can simultaneously predict all of the labels used to annotate an image. Moreover, we constructed ensembles (random forests) of PCTs, to improve the predictive performance. We tested our system on the image database from the ImageCLEF@ICPR 2010 photo annotation task. The extensive experiments conducted on the benchmark database show that our system has very high predictive performance and can be easily scaled to large number of visual concepts and large amounts of data. © 2010 Springer-Verlag.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Detection of visual concepts and annotation of images using predictive clustering trees

Detection of visual concepts and annotation of images using ...

引用

2010 Cross Language Evaluation Forum Conference, CLEF 2010

作者： Dimitrovski, Ivica Kocev, Dragi Loskovska, Suzana Džeroski, Sašo Department of Knowledge Technologies Jožef Stefan Institute Jamova cesta 39 Ljubljana1000 Slovenia Department of Computer Science Faculty of Electrical Engineering and Information Technology Karpoš bb Skopje1000 Macedonia

In this paper, we present a multiple targets classification system for visual concepts detection and image annotation. Multiple targets classification (MTC) is a variant of classification where an instance may belong to multiple classes at the same time. The system is composed of two parts: feature extraction and classification/annotation. The feature extraction part provides global and local descriptions of the images. These descriptions are then used to learn a classifier and to annotate an image with the corresponding concepts. To this end, we use predictive clustering trees (PCTs), which are capable to classify an instance to multiple classes at once, thus exploit the interactions that may occur among the different visual concepts (classes). Moreover, we constructed ensembles (random forests) of PCTs, to improve the predictive performance. We tested our system on the image database from the visual concept detection and annotation task part of ImageCLEF 2010. The extensive experiments conducted on the benchmark database show that our system has very high predictive performance and can be easily scaled to large number of images and visual concepts.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

MCRNR: Fast computing of restricted nash responses by means of sampling

MCRNR: Fast computing of restricted nash responses by means ...

引用

作者： Ponsen, Marc Lanctot, Marc De Jong, Steven Department of Knowledge Engineering Maastricht University Maastricht Netherlands Department of Computer Science University of Alberta Edmonton AB Canada Computational Modeling Lab Vrije Universiteit Brussel Brussel Belgium

ISBN: (纸本)9781577354697

This paper presents a sample-based algorithm for the computation of restricted Nash strategies in complex extensive form games. Recent work indicates that regret-minimization algorithms using selective sampling, such as Monte-Carlo Counterfactual Regret Minimization (MCCFR), converge faster to Nash equilibrium (NE) strategies than their non-sampled counterparts which perform a full tree traversal. In this paper, we show that MCCFR is also able to establish NE strategies in the complex domain of Poker. Although such strategies are defensive (i.e. safe to play), they are oblivious to opponent mistakes. We can thus achieve better performance by using (an estimation of) opponent strategies. The Restricted Nash Response (RNR) algorithm was proposed to learn robust counter-strategies given such knowledge. It solves a modified game, wherein it is assumed that opponents play according to a fixed strategy with a certain probability, or to a regret-minimizing strategy otherwise. We improve the rate of convergence of the RNR algorithm using sampling. Our new algorithm, MCRNR, samples only relevant parts of the game tree. It is therefore able to converge faster to robust best-response strategies than RNR. We evaluate our algorithm on a variety of imperfect information games that are small enough to solve yet large enough to be strategically interesting, as well as a large game, Texas Hold'em Poker. Copyright © 2010, Association for the Advancement of Artificial Intelligence. All rights reserved.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

The scaffolding of a formal context

The scaffolding of a formal context

引用

7th International Conference on Concept Lattices and Their Applications, CLA 2010

作者： Doerfel, Stephan Knowledge and Data Engineering Group Department of Mathematics and Computer Science University of Kassel Wilhelmshöher Allee 73 34121 Kassel Germany Department of Mathematics Institute of Algebra Technical University of Dresden 01062 Dresden Germany

The scaffolding of a complete lattice L of finite length was introduced by Rudolf Wille in 1976 as a relative subsemilattice of L that can be constructed using subdirect decomposition. The lattice is uniquely defined by its scaffolding and can be reconstructed from it. Using bonds, we demonstrate how the scaffolding can be constructed from a given formal context and thereby extend the notion of the scaffolding to doubly founded lattices. Further, we explain the creation of a suitable graphical representation of the scaffolding from the context.

关键词： Scaffolds

来源：评论

学校读者我要写书评

暂无评论

Intelligent hybrid architecture for tourism services

引用

作者： Almeida, Ana Coelho, Bruno Martins, Constantino GECAD Knowledge Engineering and Decision Support Group Australia Computer Science Department Institute of Engineering Polytechnic of Porto R. Dr. António Bernardino de Almeida 431 Porto 4200-072 Portugal

ISBN: (纸本)3642152856

This paper presents an artificial intelligence-based architecture to be used in various tourism services, namely the user modeling and recommender system components. Through the use of several machine learning techniques, such as linear models, neural networks, classification and even text mining, a hybrid and complete approach at understanding tourism application users is attained. To assess the advantages of the model, a specific prototype was created, including a complex user modeling process and an advanced and innovative recommender system. The system was tested in the scope of Oporto city, in Portugal. © 2010 IFIP.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Mirror my emotions! Combining facial expression analysis and synthesis on a robot

Mirror my emotions! Combining facial expression analysis and...

引用

2nd International Symposium on New Frontiers in Human-Robot Interaction - A Symposium at the AISB 2010 Convention

作者： Sosnowski, Stefan Mayer, Christoph Kühnlenz, Kolja Radig, Bernd Department of Electrical Engineering and Information Technologies Institute of Automatic Control Engineering Technische Universität München Munich Germany Computer Science Department Department for Image Understanding and Knowledge-Based Systems Technische Universität München Munich Germany Technische Universität München Munich Germany Computer Science Department Technische Universität München Munich Germany

ISBN: (纸本)1902956877

Everyday human communication relies on a large number of different communication mechanisms like spoken language, facial expressions, body pose or gestures. Facial expressions are one of the main communication mechanisms and pass large amounts of information between human dialogue partners [22]. Therefore, the analysis and the synthesis of facial expressions are important steps towards an intuitive human-machine interaction and form valuable research targets. We present a system that tackles both challenges. It relies on a fully automated, model-based, real-time capable approach to distinguish universal facial expressions and their intensities from camera images. Facial expression synthesis is conducted via the robot head EDDIE, a flexible low-cost emotion-display with 23 degrees of freedom. Static facial expressions at continuous intensities are included, as well as smooth transitions based on the circumplex model of affect. Miniature off-the-shelf mechatronic components are used to provide high functionality at low cost. Evaluations conducted in a user-study show that emotions displayed by EDDIE are recognizable by humans very well. By combining facial expression recognition and display on the robot, a first demonstration is presented in which the robot mirrors the human's emotions, as a basis for further research in the field of emotional closed loop systems.

关键词： Mirrors

来源：评论

学校读者我要写书评

暂无评论

Performance study on logging to another main memory database

Performance study on logging to another main memory database

引用

作者： Qin, Xiongpai Xiaoyun, Zhou Key Laboratory of Data Engineering and Knowledge Engineering School of Information Renmin University of China Beijing 100872 China Department of Computer Science and Technology Xuzhou Normal University Xuzhou Jiangsu 221008 China

ISBN: (纸本)9788988678183

In update intensive main memory database applications, huge volume of log records is generated, to maintain the ACID properties of the database system, the log records should be persistent efficiently. Delegating logging of one main memory database to another main memory database is proposed. The scheme is elaborated in detail in terms of architecture, logging & safeness levels, checkpointing, and recovery. Strict durability and relax durability are provided. When some form of non-volatile memory is used to temporarily holding log records, not only logging efficiency is improved, but also the scheme could guarantee full ACID of the system. We also propose using parallel logging to speedup log persistence by writing logs to multiple disks in parallel. Since interconnection network techniques progress by leaps and bounds, the scheme eliminates the concern about whether the system's overall performance may be slowed down by bandwidth and latency limitations. Experiment results demonstrate the feasibility of the proposal.

关键词： Interconnection networks (circuit switching)

来源：评论

学校读者我要写书评

暂无评论

A diversity-guided heuristic-based genetic algorithm for triangulation of Bayesian networks

A diversity-guided heuristic-based genetic algorithm for tri...

引用

作者： Dong, Xuchu Yu, Haihong Ouyang, Dantong Ye, Yuxin Zhang, Yonggang Department of Computer Science and Technology Jilin University Changchun Jilin 130012 China Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of Education Changchun Jilin 130012 China

ISBN: (纸本)9788988678251

For the optimization problem about triangulation of Bayesian networks, a novel genetic algorithm, DHGA, is proposed in this paper. DHGA employs a heuristic-based mutation operation. Moreover, it uses population diversity to identify stagnation and convergence as well as to guide the search procedure. Experiments on representative benchmarks show that DHGA posses better performance and robustness than other swarm intelligence methods.

关键词： Bayesian networks

来源：评论

学校读者我要写书评

暂无评论

An analysis of current trends in CBR research using multiview clustering

An analysis of current trends in CBR research using multivie...

引用

作者： Greene, Derek Freyne, Jill Smyth, Barry Cunningham, Pádraig Clique Research Cluster School of Computer Science and Informatics University College Dublin Ireland CSIRO's ICT Centre Based Hobart Australia Computer Science School of Computer Science and Informatics University College Dublin Ireland ChangingWorlds Ltd. Ireland Clarity: The Centre for Sensor Web Technologies Ireland Department of Knowledge and Data Engineering School of Computer Science and Informatics University College Dublin Ireland

The European Conference on Case-Based Reasoning (CBR) in 2008 marked IS years of international and European CBR conferences where almost seven hundred research papers were published. In this report we review the research themes covered in these papers and identify the topics that are active at the moment. The main mechanism for this analysis is a clustering of the research papers based on both cocitation links and text similarity. It i.ç interesting to note that the core set of papers has attracted citations from almost three thousand papers outside the conference collection so it is clear that the CBR conferences are a subpart of a much larger whole. It is remarkable that the research themes revealed by this analysis do not map directly to the subtopics of CBR that might appear in a textbook. Instead they reflect the applications-oriented focus of CBR research, and cover the promising application areas and research challenges that are faced. Copyright © 2010, Association for the Advancement of Artificial Intelligence. All rights reserved.

关键词： Case based reasoning

来源：评论

学校读者我要写书评

暂无评论

Mining and tracking evolving web user trends from large web server logs

引用

Statistical Analysis and Data Mining 2010年第2期3卷 106-125页

作者： Hawwash, Basheer Nasraoui, Olfa Knowledge Discovery and Web Mining Laboratory Department of Computer Engineering and Computer Science University of Louisville Louisville KY 40292 United States

Recently, online organizations became interested in tracking users' behavior on their websites to better understand and satisfy their needs. In response to this need, web usage mining tools were developed to help them use web logs to discover usage patterns or profiles. However, since website usage logs are being continuously generated, in some cases, amounting to a dynamic data stream, most existing tools are still not able to handle their changing nature or growing size. This paper proposes a scalable framework that is capable of tracking the changing nature of user behavior on a website, and represent it in a set of evolving usage profiles. These profiles can offer the best usage representation of user activity at any given time, and they can be used as an input to higher-level applications such as a web recommendation system. Our specific aim is to make the hierarchical unsupervised niche clustering (HUNC) algorithm more scalable, and to add integrated profile tracking and cluster-based validation to it. Our experiments on real web log data confirm the validity of our approach for large data sets that previously could not be handled in one shot. © 2010 Wiley Periodicals, Inc.

关键词： Websites

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：