检索结果-内蒙古大学图书馆

International Conference on machine learning and Cybernetics

作者： He, Qing Li, Ning Shi, Zhong-Zhi Key Laboratory of Intelligent Information Processing Institute of Computing Technology Hebei University Baoding 071002 Hebei China Graduate University of Chinese Academy of Sciences Hebei University Baoding 071002 Hebei China Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei China

ISBN: (纸本)9781424465262

Support Vector machine (SVM) is a classification technique of machine learning based on statistical learning theory. A quadratic optimization problem needs to be solved in the algorithm, and with the increase of the samples, the time complexity will also increase. So it is necessary to shrink training sets to reduce the time complexity. A sample selection method for SVM is proposed in this paper. It is inspired from the Hyper surface classification (HSC), which is a universal classification method based on Jordan Curve Theorem, and there is no need for mapping from lower-dimensional space to higher-dimensional space. The experiments show that the algorithm shrinks training sets keeping the accuracy for unseen vectors high. © 2010 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

A personalized products selection assistance based on e-commerce machine learning

A personalized products selection assistance based on e-comm...

引用

Proceedings of 2004 International Conference on machine learning and Cybernetics

作者： Yang, Hong-Wei Pan, Zhi-Geng Wang, Xi-Zhao Xu, Bing China State Key Lab. of CAD and CG Zhejiang University Hangzhou 310027 China Machine Learning Center Fac. of Math. and Computer Science Hebei University Baoding 071002 China

ISBN: (纸本)0780384032

The prosperity of electronic commerce has changed the traditional trading behaviors. More and more people are willing to perform Internet shopping. At the same time, consumers experience information overload and look for help in selecting from an overwhelming array of products. In order to overcome such a problem one option is to develop a personalized online assistance to retrieve product information that really matters for the customers. In this paper, we present a method that combines the genetic algorithm and k nearest neighbor technology to reason about the customer's personal preferences from his/her profile and then provide the most appropriate products to meet his/her needs. Our experimental results are showing that our systems have a bright future.

关键词： Electronic commerce

来源：评论

学校读者我要写书评

暂无评论

A deep web interface integration approach based on domain knowledge and submitting queries

引用

Journal of Information and computational Science 2009年第3期6卷 1103-1111页

作者： Yuan, Fang Liu, Hongfei Wei, Yonggang Yao, Zengli College of Mathematics and Computer Science Hebei University Hebei Baoding 071002 China Key Lab. in Machine Learning and Computational Intelligence of Hebei Province Hebei Baoding 071002 China

Deep Web can provide us a great amount of high quality information. In order to make full use of the information, it is becoming urgent to establish Deep Web data integration system, in which Deep Web interface integration is an important component. This paper proposes a deep web interface integration method based on domain knowledge and submitting queries for certain domains. Before application this method, we categorize Deep Web interfaces according to their structures and introduce the concept of interface element, which be used to represent the interface. This approach first selects the required attributes, and then matches these attributes in each interface element, after that integrates each interface element which matches the same attribute to generate the global interface. This method includes interface template matching, domain knowledge-based interface elements matching and submitting queries-based interface elements matching, etc. Experimental results show that the proposed method has higher matching accuracy and lower dependency on preparatory work. 1548-7741/ Copyright © 2009 Binary Information Press.

关键词： Domain Knowledge

来源：评论

学校读者我要写书评

暂无评论

Incrementally Updating Method in Dominance- Based Rough Set Approach

Incrementally Updating Method in Dominance- Based Rough Set ...

引用

作者： Yan LI Xiaoqing LIU Jiajia HOU Key Lab.In Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University

Dominance-based rough set approach(DRSA) can handle the attributes with preference orders, and therefore it has been widely applied in multi-criteria decision making problems. In real applications, the collected information is updated from time to time which results in dynamic information systems, especially when the attributes or objects are inserted or deleted. The traditional DRSA needs to update the set approximations whenever the information systems change, which decreases the method efficiency greatly. For classification problems with multiple criteria, this paper presents incremental algorithms to update set approximations when an object is inserted or deleted, which is expected to be more efficient than computing the approximations from the scratch. The related theoretical results are presented with proofs, and illustrative examples are also given to support the effectiveness of the proposed incremental method.

关键词： Rough Set Dominance Relation Set Approximation Dynamic Information Systems Incremental Update

来源：评论

学校读者我要写书评

暂无评论

Updating Approximations of VPRS Model based on Dominance Relations

Updating Approximations of VPRS Model based on Dominance Rel...

引用

作者： Jiajia HOU Yan LI Yan ZHAN Liang WANG Key Lab.In Machine Learning and Computational Intelligence College of Mathematics and Computer ScienceHebei University

In real world problems, the collected data vary from time to time, and therefore, the approximations of a concept by a variable precision rough set model(VPRS) should be correspondingly updated. This paper focuses on developing incremental method to update set approximations of VPRS based on dominance relations. Under dynamic environments where an object is inserted or deleted, we present the updating principles and then develop the incremental method for updating approximation sets. The related theoretical results are presented with proofs, and illustrative examples are also given to support the effectiveness of the proposed method.

关键词： Variable Precision Rough Set Dominance Relation Set Approximation Dynamic Information Systems Incremental Update

来源：评论

学校读者我要写书评

暂无评论

A Core Set Weighted Support Vector machines

A Core Set Weighted Support Vector Machines

引用

作者： Shuxia LU Limin LI Key Lab.In Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University

Classification for large datasets is a classical problem in machine learning. In this paper, we focus on effevtive classification algorithm for large datasets and imbalanced datasets. First, to deal with imbalanced dataset, we define the weight according to the size of positive and negative dataset. Then, a fast learning algorithm on large datasets called a core set weighted support vector machines(CSWSVM) is proposed. In the proposed approach, the corresponding core set(CS) can be solved by employing the core vector machine(CVM) or generalized CVM(GCVM), and then the weighted support vector machines(WSVM) can be used to implement classification for imbalanced datasets. Experimental results on UCI and USPS datasets demonstrate that the proposed method is effective.

关键词： core vector machine weight support vector machine core set

来源：评论

学校读者我要写书评

暂无评论

Variable-precision Rough Set Approach based on Extended Dominance Relations

Variable-precision Rough Set Approach based on Extended Domi...

引用

作者： Yan LI Yongfei JIN Key Lab.Of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University

According to the definition of dominance relation, an object x is said to dominate another object y only when x dominates y on all attributes, which is too strict especially when the number of attributes is large. To cope with this problem, the extended dominance-based rough set model has been developed by introducing a parameter to the concept of traditional dominance relationship in the reported literature. However, in this extended model, the definitions of lower and upper approximations are the same to the traditional model, which may affect the decision making process. In this paper, we introduce the idea of variable precision to the extended dominance rough set model for better fault tolerance ability. The impact of the parameter on decision results with respect to testing accuracy is studied. Finally, an example is given and the experimental results on UCI data are also shown to support the effectiveness of the proposed method.

关键词： Extended dominance relation Variable precision rough set Dominant sets Approximation sets

来源：评论

学校读者我要写书评

暂无评论

A Fast Algorithm for Computing Dominance Classes

A Fast Algorithm for Computing Dominance Classes

引用

作者： Yan LI Qun YU Key Lab.Of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University

Traditional rough set theory(TRS) is based on the concept of equivalence relation to define upper and lower approximation sets of a given target concept, and therefore uncertainties in information systems can be represented. By using equivalence relations, TRS only considers whether attribute values are distinguished or not, regardless of the preference information contained in attribute values. Rough sets based on dominance relations effectively solve this problem and can deal with preference-ordered data. In these dominance-based approaches, the computational cost of the dominance classes greatly affects the efficiency of attribute reduction and rule extraction. This paper presents an efficient method of computing dominance classes in an ordered information system by rapidly reducing the search space. Based on the definition of dominance class, the inferior class of an object is gradually removed from the universe with the increase of the attributes in the computation process. Experiments on ten UCI data sets show that the proposed algorithm obviously improves the efficiency of computing dominance classes, especially for large-scale data.

关键词： Rough set dominance class ordered information systems fast algorithm

来源：评论

学校读者我要写书评

暂无评论

A Rough Set Model Based on Extended Dominance Relations

A Rough Set Model Based on Extended Dominance Relations

引用

作者： Yan LI Shuai SUN Yongfei JIN Yan ZHAN Key Lab.In Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University

Dominance relation rough set approach(DRSA) is a useful mathematical tool to deal with preference-ordered data. The main idea is using dominance relations to replace equivalent relations in classical rough set theory. However, the definition of conventional dominance relation is very strict which may limit its application to information systems with relative large number of attributes. In this paper, we relax the conditions in the definition of dominance relation and introduce the concept of extended dominance relation. The proprieties of this new concept are also discussed and it is found that all the properties of classical dominance relation are still satisfied.

关键词： Rough Set Information System Dominance Relation DRSA Extended Dominance Relation

来源：评论

学校读者我要写书评

暂无评论

A Multi-task Pathfinding Method Based on CBR and Kd-tree

A Multi-task Pathfinding Method Based on CBR and Kd-tree

引用

作者： Yan Li Sen Liang Lanming Su Jiacheng He Key Lab.In Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University

Pathfinding is a typical task in many computer games, and its performance will affect the quality of game AI. In order to enhance the efficiency of multi-task pathfinding, case-based reasoning has been introduced in traditional A* algorithm, called the CBMT method. The method needs to select representative paths which can cover the whole map to build a compact case base, which is difficult in large maps. Besides, repeatedly searching for similar cases for each pathfinding task would be a time consuming process. To address these problems, we provide a kd-tree case storage structure and case retrieval mechanical in the CBMT method. The pre-stored cases(previously found paths) are generated randomly and incrementally. The original flat storage structure of the cases is changed into the kd-tree structure. Since the searching space can be reduced by branch pruning in case retrieval, the pathfinding efficiency has been improved obviously, and the number of searched nodes is also reduced.

关键词： Multi-task Pathfinding Case-based Reasoning Kd-tree Nearest-neighbors Search Algorithm A

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：