【摘 要】
：
The learning of Credit Scoring has recently gained much attention,and many methods based on machine learning approaches have been proposed.Based on the above research,most of existing Credit Scoring m
【机 构】
：
Canvard College,Beijing Technology and Business University,Beijing 101118,China
【出 处】
：
2014年国际计算机科学与软件工程学术会议
论文部分内容阅读
The learning of Credit Scoring has recently gained much attention,and many methods based on machine learning approaches have been proposed.Based on the above research,most of existing Credit Scoring methods take vectors as their input data,and then a function is learned in such a vector space for classification,clustering,or dimensionality reduction.However,in some cases,there are some reasons to take tensors as their input data,e.g.,an image can be considered as a second order tensor.It is reasonable to consider that pixels close to each other are correlated to some extent.In this paper,we represent the data points by second order tensors rather than vectors,and then establish a new Credit Scoring model,which based on support tensor machine.To solve this model,an iterative algorithm is used.This tensor-based algorithm requires a smaller set of decision variables as compared to vector-based approaches,thus,it is helpful to overcome small-sample-size problems in vector-based learning.We compare our proposed method with SVM and LS-SVM on German credit Dataset.Experimental results show the effectiveness of our method.
其他文献
PID controller is the most widely used controller.The parameters of traditional PID controller are fixed,and the scope of adapting is limited.In this paper,the RBF neural network controller is used to
At present,the quantitative analysis and empirical research of Chinese academia on government population scale was limited to two basic perspectives,the vertical and horizontal.Quantitative study of t
Community structure is an important property to uncover structural and functional features in various complex systems.In this paper,we propose an improved spread algorithm based on Principal Component
Fault localization techniques are promising in reducing debugging efforts.To effectively locate faults,many fault localization techniques require a large number of test cases whose execution results a
As many network applications attackers may utilize the protocol vulnerabilities for spreading malicious codes,or exploit some unknown protocols to transfer data secretly.The techniques of inferring pr
This paper proposes a new algorithm based on Support Vector Data Description which labeling samples automatically by lead transductive confidence.The proposed algorithm gives the confidence lever abou
In this paper,we discuss the exact inference of the scale parameter of exponential distribution with consideration of Type-Ⅰ hybrid censoring scheme.Using the Weibull-to-exponential transformation,we
For detection of magnetic induction coil section only able to show the status of vehicle information,proposed intersection signal control to design a sensor-based electronic license plate.The design i
UML is accepted as standard modeling language by the public in the software development.Using UML model transformation can accelerate the development progress of software and guarantee the consistency
There are various of materials science data resources in steel domain,and most of these open data resources are available.However,these open data resources may reside in different web sites,literature