检索结果-南通市图书馆

The new interpretation of support vector machines on statistical learning theory

science china Mathematics 2010年第1期53卷 151-164页

作者： ZHANG ChunHua 1 , TIAN YingJie 2 & DENG NaiYang 3,1 School of Information, Renmin University of china, beijing 100872, china 2 research center on fictitious economy and data science, chinese academy of sciences, beijing 100080, china 3 College of science, china Agricultural University, beijing 100083, china 1. School of Information Renmin University of China Beijing 100872 China2. Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing 100080 China3. College of Science China Agricultural University Beijing 100083 China

This paper is concerned with the theoretical foundation of support vector machines (SVMs). The purpose is to develop further an exact relationship between SVMs and the statistical learning theory (SLT). As a representative, the standard C-support vector classification (C-SVC) is considered here. More precisely, we show that the decision function obtained by C-SVC is just one of the decision functions obtained by solving the optimization problem derived directly from the structural risk minimization principle. In addition, an interesting meaning of the parameter C in C-SVC is given by showing that C corresponds to the size of the decision function candidate set in the structural risk minimization principle.

关键词： C-support vector classification the minimization principle of the structural risk KKT conditions

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Regional differences and determinants of built-up area expansion in china

引用

science china Earth sciences 2007年第12期50卷 1835-1843页

作者： HUANG JiKun1,2,ZHU LiFen3,1 & DENG XiangZheng1,2 1 center for chinese Agricultural Policy,chinese academy of sciences,beijing 100101,china 2 Institute of Geographic sciences and Natural Resources research,chinese academy of sciences,beijing 100101,china 3 research center of the economy of the Upper Reaches of the Yangtze River,Chongqing Technology and Business University,Chongqing 400067,china 1. center for chinese Agricultural Policy Chinese Academy of Sciences Beijing 100101 China 2. Institute of Geographic sciences and Natural Resources research Chinese Academy of Sciences Beijing 100101 China 3. research center of the economy of the Upper Reaches of the Yangtze River Chongqing Technology and Business University Chongqing 400067 China

Based on remote sensing data on land use provided by the chinese academy of sciences and socio-economic data collected by the authors,this paper analyzes the trends and regional differences in built-up area(BUA) expansion in china from the late 1980s to 2000,and empirically estimates the major determinants of BUA expansion in different regions in 1996―*** 1989―2000,although china's overall BUA expansion accelerated,the trends differed significantly among *** expansion in the central and western regions accelerated significantly,but it slowed down considerably in the east-ern *** estimation results from our econometric analysis reveal that BUA expansion in the eastern region reached a period when economic growth had no further significant impact on per capita BUA,the land utilization in this region has become more intensive with further expansion of the *** the central and western regions,the BUA has expanded remarkably due to the relatively more flexible land development policies and the relatively cheap land ***,as the econ-omy continues to grow rapidly,policies relating to BUA expansion and cultivated land reductions may face more serious challenges in the central and western regions.

关键词： built-up area, land use, regional difference, determinants, china

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

CO_2 Emission Induced by Urban Household Consumption in china

引用

chinese Journal of Population,Resources and Environment 2009年第3期7卷 11-19页

作者： Wang Yan1,2,3,Shi Minjun1,2 1.Graduate University of chinese academy of sciences,beijing 100049,china 2.research center on fictitious economy and data science,CAS,beijing 100190,china 3.Communications University of china,beijing 100024,china Graduate University of chinese academy of sciences research center on fictitious economy and data science CAS Communications University of china

Household consumption is one of the important factors that induce COL emission. Based on input-output model, this article calculated the intensity of CO2 emission of different income groups and seven provinces in china, and then estimated total CO2 emission induced by urban household consumption from 1995 to 2004 in china based on statistic data of household living expenditure. The results show that CO2 emission per capita induced by household consumption had increased from 1583 to 2498 kg CO2 during 1995-2004. The ratio of consumption-induced CO2 emission to total CO2 emission had risen from 19% to 30% in the past decade. Indirect CO2 emission accounted for an important part of the consumption-induced emission, the ratio of indirect emission to consumption-induced emission had risen from 69% to 79% during the same period. A significant difference in consumption-induced CO2 emission across different income groups and different regions has been observed. COs emission per capita of higher income groups and developed regions increased faster than that of lower income groups and developing regions. Changing lifestyle has driven significant increase in CO2 emission. Especially, increases in private transport expenditure （for example, vehicle expenditure） and house building expenditure are key driving factors of growth in consumption-induced COL emission. There are big differences in the amount of CO2 emission induced by change in lifestyle across different income groups and provinces. It can be expected that lower income households and developing regions will increase consumption to improve their livings with income growth in the future, which may induce much more CO2 emission. A reasonable level of CO2 emission is necessary to satisfy human needs and to improve living standard, but a noticeable fact is that CO2 emission per capita induced by household consumption in developed areas of china had reached a quite high level. Adjustment in lifestyle towards a low-carbon soc

关键词： CO2 emission urban household consumption lifestyle input-output model china

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Application on the Effect of the Feature Attributes to Non-Linear Classification

Application on the Effect of the Feature Attributes to Non-L...

引用

第六届（2011）中国管理学年会——商务智能分会场

作者： Jing Chu research center on fictitious economy & data science,chinese academy of sciences,china Yong Shi research center on fictitious economy & data science,chinese academy of sciences,china

A deeper discussion on contributions from feature attributes to the classifying attribute in the nonlinear classification model based on the Choquet integral is given in this *** work provides some detail examples to show the application to the geometric structure of the model with contribution rates from the feature attributes towards the classification,especially the interaction among them.

关键词：

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

A New Decomposition Algorithm for Training Bound-Constrained Support Vector Machines

A New Decomposition Algorithm for Training Bound-Constrained...

引用

第四届（2009）中国管理学年会--系统管理与复杂性科学分会场

作者： Lingfeng NIU,Yong SHI CAS research center on fictitious economy & data science Graduate School of chinese academy of science

This work proposes a new decomposition algorithm for training bound-constrained Support Vector Machines. A new working set selection rule together with the global convergence based on this rule are given. In order to make the algorithm efficient in practice, projected gradient algorithm and interior point method are combined together to solve the quadratic subproblem. Numerical results on several public data sets validate the correctness and efficiency of the proposed algorithm.

关键词：

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

CUKNN: A PARALLEL IMPLEMENTATION OF K-NEAREST NEIGHBOR ON CUDA- ENABLED GPU

CUKNN: A PARALLEL IMPLEMENTATION OF K-NEAREST NEIGHBOR ON CU...

引用

2009 IEEE Youth Conference on Information,Computing and Telecommunication(YC-ICT2009)

作者： Shenshen Liang 1 , Cheng Wang 2 Ying Liu 1,3 , Liheng Jian 1 1 Graduate University of chinese academy of sciences, beijing, china 100190 2 Agilent Technologies Co. Ltd., beijing, china 100102 3 fictitious economy and data science research center, chinese academy of sciences beijing, china 100190

Recent development in Graphics Processing Units (GPUs) has enabled inexpensive high performance computing for general-purpose applications. Due to GPU's tremendous computing capability, it has emerged as the co-processor of the CPU to achieve a high overall throughput. CUDA programming model provides the programmers adequate C language like APIs to better exploit the parallel power of the GPU. K-nearest neighbor is a widely used classification technique and has significant applications in various domains. The computational-intensive nature of KNN requires a high performance implementation. In this paper, we present a CUDA-based parallel implementation of KNN, CUKNN, using CUDA multi-thread model. Various CUDA optimization techniques are applied to maximize the utilization of the GPU. CUKNN outperforms significantly and achieve up to 15.2X speedup. It also shows good scalability when varying the dimension of the training dataset and the number of records in training dataset.

关键词： KNN, classification, CUDA, parallel computing

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

A hybrid of fuzzy-link clustering and classification for seismic data

A hybrid of fuzzy-link clustering and classification for sei...

引用

第六届（2011）中国管理学年会——商务智能分会场

作者： Zhongbin Ouyang1,2,Yong Shi1,2,3,Le Yang1,2 (1 research center on fictitious economy and data science,CAS,beijing 100190,china) (2 Graduate University of chinese academy of sciences,beijing 100190,china) (3 College of Information science and Technology,University of Nebraska at Omaha,Omaha,NE 68118,USA)

Interpreting seismic data is an arduous and time-consuming *** cluster analysis can be helpful to analyze seismic data,most traditional clustering algorithms can hardly obtain satisfactory results since they do not take account of spatially dependencies of seismic data,which make clustering seismic data different from other clustering ***,seismic data are by nature fuzzy and noisy,which limit many popular clustering *** this paper,we present a hybrid of fuzzy-link clustering technique and classification for seismic data incorporating spatial and non-spatial information into the *** results on real seismic data show that our algorithm is robust to noise and can discover clusters with arbitrary shape.

关键词： seismic data cluster analysis classification spatial attribute non-spatial

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Effects of expanding domestic demands on regional development in china

Effects of expanding domestic demands on regional developmen...

引用

第四届（2009）中国管理学年会--城市与区域管理分会场

作者： Minjun Shi1,2, Guoxia Ma1,2, Na Li1,2, Yongna Yuan 1. research center on fictitious economy and data science (FEDS, CAS) 1,2 2. Graduate University of chinese academy of sciences (GUCAS)

Decline in export induced by the global financial crisis has inflicted heavy damages on economic growth of the coastal regions which are deeply dependent on exportation, but relatively light impacts on the inland regions which are less dependent on exportation. With implementation of the policy to enlarge domestic demands for fighting global financial crisis, there has been an obvious contribution of increasing investment to economic growth of the inland regions. Actually in the first half of 2009, the growth rate of part of the inland provinces was higher than that of the coastal provinces. Some scholars called such a phenomenon as "adverse-gradient growth". The simulation results based on china multi-regional CGE model indicate that investment in inland regions may not only promote GDP growth of the inland regions, but also contribute to economic growth of the coastal regions through GDP spillover. Furthermore, because of variation in consumption level and market accessibility between the coastal regions and the inland regions, the impetus of consumption demands to economic growth in the coastal regions is much stronger. Therefore, so-called "adverse-gradient growth" might remain at a short-term effect which is simultaneously resulted from slowdown in GDP growth of the coastal regions caused by decline in export and contribution of increase in investment of the inland regions. If there is no concomitant geographical transfer of industries from the coastal regions to the inland regions, sustainable economic growth of the inland regions and alleviation of spatial inequality seems not optimistic.

关键词： Expanding domestic demands regional development adverse-gradient growth investment growth spillover effect

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

A Fuzzy MCDM Based Approach for Hybrid Recommender Systems

A Fuzzy MCDM Based Approach for Hybrid Recommender Systems

引用

第六届（2011）中国管理学年会——商务智能分会场

作者： Yi Peng,Wenli Zhang,and Gang Kou School of Management and Economics,University of Electronic science and Technology of china,Chengdu 610054,PR china Yong Shi College of Information science & Technology,University of Nebraska at Omaha,Omaha,NE 68182,USA and research center on fictitious economy and data sciences,chinese academy of sciences,beijing 100190,china

Recommender systems are the most widely used techniques to overcome information ***,most of current techniques make recommendations based on single *** this study,we propose a new hybrid recommendation approach based on Fuzzy Multiple Criteria Decision Making （MCDM） *** MCDM methods,DEMATEL and TOPSIS are integrated with traditional collaborative filtering and content-based *** order to evaluate proposed approach,we collect data from survey questionnaire with users from ***,the largest chinese website devoted for movies,books,and music *** use oversampling technique to simulate the recommendation data set in multi-criteria *** result shows that proposed approach can improve the hit rate comparing to collaborative filtering recommendation.

关键词：

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Comparing chinese reviews based on Fuzzy PROMETHEE: A new approach to sentiment analysis

Comparing Chinese reviews based on Fuzzy PROMETHEE: A new ap...

引用

第六届（2011）中国管理学年会——商务智能分会场

作者： Jun Li,Gang Kou,Yong Zhang,Yi Peng School of Management and Economics,University of Electronic science and Technology of china,Chengdu,P.R.china,610054Yong Shi College of Information science & Technology,University of Nebraska at Omaha,Omaha,NE 68182,USA and research center on fictitious economy and data sciences,chinese academy of sciences,beijing 100190,china

There are now numerous product reviews provided on the Internet,which makes it difficult for a potential customer to make an informed decision on whether to purchase the *** different products and services are alternatives to choose,and various product features can be explicitly considered as multi-criteria,product comparison can also be a multi-criteria decision making *** this paper,we aim to introduce an MCDM technology to compare consumer opinions at the level of product features and give a ranking *** ultimate result using 4 types of mobile phone reviews demonstrated the potential effectiveness of this method.

关键词： sentiment analysis opinion mining MCDM fuzzy PROMETHEE

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：