• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.3
• /
• pp.263-267
• /
• 2005

Building classifiers for financial real-world classification problems is often plagued by severely overlapping and highly skewed class distribution. New performance measures such as receiver operating characteristic (ROC) curve and area under ROC curve (AUC) have been recently introduced in evaluating and building classifiers for those kind of problems. They are, however, in-effective to evaluation of classifier's discrimination performance in a particular class of the classification problems that interests lie in only a local operating range of the classifier, In this paper, a new method is proposed that enables us to directly improve classifier's discrimination performance at a desired local operating range by defining and optimizing a partial area under ROC curve or domain-specific curve, which is difficult to achieve with conventional classification accuracy based learning methods. The effectiveness of the proposed approach is demonstrated in terms of fraud detection capability in a real-world fraud detection problem compared with the MSE-based approach.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.1
• /
• pp.59-64
• /
• 2010

To improve system performance, we apply the concept of membership function to Variance Considered Machines (VCMs) which is a modified algorithm of Support Vector Machines (SVMs) proposed in our previous studies. Many classification algorithms separate nonlinear data well. However, existing algorithms have ignored the fact that probabilities of error are very high in the data-mixed area. Therefore, we make our algorithm ignore data which has high error probabilities and consider data importantly which has low error probabilities to generate system output according to the probabilities of error. To get membership function, we calculate sigmoid function from the dataset by considering means and variances. After computation, this membership function is applied to the VCMs.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.191-194
• /
• 2002

In this paper, a new GA-based methodology with information granules is suggested for construction of the fuzzy classifier. We deal with the selection of the fuzzy region as well as two major classification problems-the feature selection and the pattern classification. The proposed method consists of three steps: the selection of the fuzzy region, the construction of the fuzzy sets, and the tuning of the fuzzy rules. The genetic algorithms (GAs) are applied to the development of the information granules so as to decide the satisfactory fuzzy regions. Finally, the GAs are also applied to the tuning procedure of the fuzzy rules in terms of the management of the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example-the classification of the Iris data, is provided.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.7 no.3
• /
• pp.165-170
• /
• 2007

Recommendation systems provide users with proper services using context information being input from many sensors occasionally under ubiquitous computing environment. But in case there isn't sufficient context information for service recommendation in spite of much context information, there can be problems of resulting in inexact result. In addition, in the quantification step to use context information, there are problems of classifying context information inexactly because of using an absolute classification course. In this paper, we solved the problem of lack of necessary context information for service recommendation by using dynamic profile information. We also improved the problem of absolute classification by using a relative classification of context information in quantification step. As the result of experiments, expectation preference degree was improved by 7.5% as compared with collaborative filtering methods using an absolute quantification method where context information of P2P mobile agent is used.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.3
• /
• pp.202-209
• /
• 2002

This paper deals with simplification of classification rules for data mining and rule bases for control systems. Datamining that extracts useful information from such a large amount of data is one of important issues. There are various ways in classification methodologies for data mining such as the decision trees and neural networks, but the result should be explicit and understandable and the classification rules be short and clear. The rough sets theory is an effective technique in extracting knowledge from incomplete and inconsistent data and provides a good solution for classification and approximation by using various attributes effectively This paper investigates granularity of knowledge for reasoning of uncertain concopts by using rough set approximations and uses a hierarchical classification structure that is more effective technique for classification by applying core to upper level. The proposed classification methodology makes analysis of an information system eary and generates minimal classification rules.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.360-366
• /
• 2008

The initial fuzzy partitions in fuzzy rule-based classification systems are determined by considering the domain region of each attribute with the given data, and the optimal classification boundaries within the fuzzy partitions can be discovered by tuning their parameters using various learning processes such as neural network, genetic algorithm, and so on. In this paper, we propose a selection method for fuzzy partition based on statistical information to maximize the performance of pattern classification without learning processes where statistical information is used to extract the uncertainty regions (i.e., the regions which the classification boundaries in pattern classification problems are determined) in each input attribute from the numerical data. Moreover the methods for extracting the candidate rules which are associated with the partition intervals generated by statistical information and for minimizing the coupling problem between the candidate rules are additionally discussed. In order to show the effectiveness of the proposed method, we compared the classification accuracy of the proposed with those of conventional methods on the IRIS and New Thyroid Cancer data. From experimental results, we can confirm the fact that the proposed method only considering statistical information of the numerical patterns provides equal to or better classification accuracy than that of the conventional methods.

• Journal of Intelligence and Information Systems
• /
• v.1 no.1
• /
• pp.61-72
• /
• 1995

Dynamic and explicit ordering of strategies is a key process in modeling knowledge-level problem-solving behavior. This paper addressed the important problem of howl to make the scheduler more knowledge-intensive in a way that facilitates the acquisition, integration, and maintenance of the scheduler control knowledge. The solution a, pp.oach described in this paper involved formulating the scheduler task as a heuristic classification problem, and then implementing it as a classification expert system. By doing this, the wide spectrum of known methods of acquiring, refining, and maintaining the knowledge of a classification expert system are a, pp.icable to the scheduler control knowledge. One important innovation of this research is that of recursive heuristic classification : this paper demonstrates that it is possible to formulate and solve a key subcomponent of heuristic classification as heuristic classification problem. Another key innovation is the creation of a method of dynamic heuristic classification : the classification alternatives that are selected among are dynamically generated in real-time and then evidence is gathered for and aginst these alternatives. In contrast, the normal model of heuristic classification is that of structured selection between a set of preenumerated fixed alternatives.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.15-18
• /
• 2000

인터넷을 포함한 여러 가지 기술이 발전됨에 따라 인터넷 광고 시장도 두드러지게 성장하고 있다. 무한한 가상 공간과 디지털이 갖는 신속성, 편리함 등 다양한 서비스를 제공할 수 있는 인터넷 기반의 모든 서비스는 데이터베이스로 축적된다. 여기에 애매함과 불확실성을 가지는 실세계를 표현하기 위해 퍼지논리함수를 이용하여 효율적으로 방문객을 분류하는 방법을 제시한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.281-285
• /
• 2002

The observation fuzzy data in random effect and balanced designs for one way classification by using a the matrix formulation, we can estimate the fuzzy variance components for the ozone depletion example and test by the agreement index.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1301-1304
• /
• 1993

The base of proposed decomposing approach is multilevel process of agregation (simplificative transformation) of the description of the project structures. The new classification of fuzzy choice operators is suggested to obtain the decomposing correlations.