• Asia pacific journal of information systems
• /
• 제20권2호
• /
• pp.23-37
• /
• 2010

Conventional Support Vector Machines (SVMs) have been utilized as classifiers for binary classification problems. However, certain real world problems, including corporate bond rating, cannot be addressed by binary classifiers because these are multi-class problems. For this reason, numerous studies have attempted to transform the original SVM into a multiclass classifier. These studies, however, have only considered nominal classification problems. Thus, these approaches have been limited by the existence of multiclass classification problems where classes are not nominal but ordinal in real world, such as corporate bond rating and multiclass customer classification. In this study, we adopt a novel multiclass SVM which can address ordinal classification problems using ordinal pairwise partitioning (OPP). The proposed model in our study may use fewer classifiers, but it classifies more accurately because it considers the characteristics of the order of the classes. Although it can be applied to all kinds of ordinal multiclass classification problems, most prior studies have applied it to finance area like bond rating. Thus, this study applies it to a real world customer level classification case for implementing customer relationship management. The result shows that the ordinal multiclass SVM model may also be effective for customer level classification.

• Journal of the Korean Data and Information Science Society
• /
• 제26권6호
• /
• pp.1557-1563
• /
• 2015

Multiclass classification is typically performed using the voting scheme method based on combining binary classifications. In this paper we propose multiclass classification method for large data, which can be regarded as the revised one-vs-all method. The multiclass classification is performed by using the hat matrix of least squares support vector machine (LS-SVM) ensemble, which is obtained by aggregating individual LS-SVM trained on each subset of whole large data. The cross validation function is defined to select the optimal values of hyperparameters which affect the performance of multiclass LS-SVM proposed. We obtain the generalized cross validation function to reduce computational burden of cross validation function. Experimental results are then presented which indicate the performance of the proposed method.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.647-650
• /
• 1999

In this Paper, we propose an optimal feature extraction for multiclass problems through proper choice of initial feature vectors. Although numerous feature extraction algorithms have been proposed, those algorithms are not optimal for multiclass problems. Recently, an optimal feature extraction algorithm for multiclass problems has been proposed, which provides a better performance than the conventional feature extraction algorithms. In this paper, we improve the algorithm by choosing good initial feature vectors. As a result, the searching time is significantly reduced. The chance to be stuck in a local minimum is also reduced.

• Communications for Statistical Applications and Methods
• /
• 제19권5호
• /
• pp.655-662
• /
• 2012

Classification is an important research field in pattern recognition with high-dimensional predictors. The support vector machine(SVM) is a penalized feature selector and classifier. It is based on the hinge loss function, the non-convex penalty function, and the smoothly clipped absolute deviation(SCAD) suggested by Fan and Li (2001). We developed the algorithm for the multiclass SVM with the SCAD penalty function using the local quadratic approximation. For multiclass problems we compared the performance of the SVM with the $L_1$, $L_2$ penalty functions and the developed method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제6권4호
• /
• pp.331-334
• /
• 2006

Original Support Vsctor Machines (SVMs) by Vapnik were used for binary classification problems. Some researchers have tried to extend original SVM to multiclass classification. However, their studies have only focused on classifying samples into nominal categories. This study proposes a novel multiclass SVM model in order to handle ordinal multiple classes. Our suggested model may use less classifiers but predict more accurately because it utilizes additional hidden information, the order of the classes. To validate our model, we apply it to the real-world bond rating case. In this study, we compare the results of the model to those of statistical and typical machine learning techniques, and another multi class SVM algorithm. The result shows that proposed model may improve classification performance in comparison to other typical multiclass classification algorithms.

• Communications for Statistical Applications and Methods
• /
• 제15권3호
• /
• pp.441-450
• /
• 2008

In this paper we propose a new method for solving multiclass problem with least squares support vector machine(LS-SVM) regression. This method implements one-against-all scheme which is as accurate as any other approach. We also propose cross validation(CV) method to select effectively the optimal values of hyper-parameters which affect the performance of the proposed multiclass method. Experimental results are then presented which indicate the performance of the proposed multiclass method.

• 전자공학회논문지
• /
• 제50권6호
• /
• pp.238-245
• /
• 2013

제스처 인식은 자연스러운 사용자 인터페이스를 위해 활발히 연구되는 중요한 분야이다. 본 논문에서는 키넥트 카메라로부터 입력되는 사용자의 3차원 관절(joint) 정보를 해석하여 제스처를 인식하는 방법이 제안된다. 대상으로 하는 제스처의 분포 특성에 따라 분류 트리를 설계하고 입력 패턴을 분류한다. 그리고 제스처를 리샘플링 및 정규화 하여 일정한 구간으로 나누고 각 구간의 체인코드 히스토그램을 추출한다. 트리의 각 노드별로 분류된 제스처에 다중 클래스 SVM(Multiclass Support Vector Machine)를 적용하여 학습한다. 이후 입력 데이터를 구성된 트리로 분류한 다음, 학습된 다중 클래스 SVM을 적용하여 제스처를 분류한다.

• 한국인공지능학회지
• /
• 제9권1호
• /
• pp.15-20
• /
• 2021

This paper was conducted to prevent and respond to crimes by predicting crimes based on artificial intelligence. While the quality of life is improving with the recent development of science and technology, various problems such as poverty, unemployment, and crime occur. Among them, in the case of crime problems, the importance of crime prediction increases as they become more intelligent, advanced, and diversified. For all crimes, it is more critical to predict and prevent crimes in advance than to deal with them well after they occur. Therefore, in this paper, we predicted crime types and crime tools using the Multiclass Logistic Regression algorithm and Multiclass Neural Network algorithm of machine learning. Multiclass Logistic Regression algorithm showed higher accuracy, precision, and recall for analysis and prediction than Multiclass Neural Network algorithm. Through these analysis results, it is expected to contribute to a more pleasant and safe life by implementing a crime prediction system that predicts and prevents various crimes. Through further research, this researcher plans to create a model that predicts the probability of a criminal committing a crime again according to the type of offense and deploy it to a web service.

• 대한교통학회지
• /
• 제19권1호
• /
• pp.77-88
• /
• 2001

현재까지의 관측교통량기반 수요추정법은 단일차종(singleclass)기반 연구가 대부분을 차지하고 있다. 그러나 현실 교통망에서는 여러 차종이 혼재되어 교통수요나 흐름을 만든다. 즉, 기존의 관측교통량기반 수요추정법은 PCE(Passenger Car Equivalent) 환산을 통한 여러 개의 차종O/D 및 관측교통량을 승용차 단위로 전환하여 하나의 O/D 및 관측교통량으로 만들어 O/D를 추정하고, 최초의 PCE환산이전 차종별 O/D의 고정비율과 관측교통량 고정비율로 곱해 차종별 O/D 및 관측교통량으로 나누어 분석하는 것이 일반적인 방법이었다. 즉, 다차종기반분석법은 각각의 차종별 O/D에 대한 노선선택비율을 각각 계산하고, 그에 따른 목적함수 감소방향인 gradient를 또한 각각 계산하여 차종별 추정력을 극대화하는 것이 그 장점이라고 할 수 있다. 따라서, 본 연구에서는 단일차종기반추정법을 다차종기반추정법으로 확장하여 차종간 혼잡을 고려한 보다 현실적인 수요추정기법을 마련하는 것이 본 연구의 목적이라고 하겠다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.643-646
• /
• 1999

In this paper, we propose a feature extraction method based on Bhattacharyya distance for multiclass problems. The Bhattacharyya distance provides a valuable information in determining the effectiveness of a feature set and has been used as separability measure for feature selection. Recently, a feature extraction algorithm hat been proposed for two normally distributed classes based on Bhattacharyya distance. In this paper, we propose to expand the previous approach to multiclass cases. Experiment results show that the proposed method compares favorably with the conventional methods.