• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.7
• /
• pp.1114-1124
• /
• 2015

In this paper, precipitation / non-precipitation pattern classification of meteorological radar data is conducted by using neuro-fuzzy algorithm. Structure expression of meteorological radar data information is analyzed in order to effectively classify precipitation and non-precipitation. Also diverse input variables for designing pattern classifier could be considered by exploiting the quantitative as well as qualitative characteristic of meteorological radar data information and then each characteristic of input variables is analyzed. Preferred pattern classifier can be designed by essential input variables that give a decisive effect on output performance as well as model architecture. As the proposed model architecture, neuro-fuzzy algorithm is designed by using FCM-based radial basis function neural network(RBFNN). Two parts of classifiers such as instance classifier part and echo classifier part are designed and carried out serially in the entire system architecture. In the instance classifier part, the pattern classifier identifies between precipitation and non-precipitation data. In the echo classifier part, because precipitation data information identified by the instance classifier could partially involve non-precipitation data information, echo classifier is considered to classify between them. The performance of the proposed classifier is evaluated and analyzed when compared with existing QC method.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.

• Journal of Ship and Ocean Technology
• /
• v.5 no.2
• /
• pp.1-13
• /
• 2001

Data in one Product Model for shipbuilding is inevitably referenced and manipulated during each phase of design or production activities, and data or manipulation status naturally varies from the original with the advance of each activities. For Object-Oriented approach, it is hard to identify classes dealing with those variations, and even if once a product model is developed, it might be getting much harder to modify it to cope with a new additional phase of activities. This paper proposes an Instance-Oriented Modeling Method, temporarily named “Concept-Relationship Modeling Approach”, which handles Data structure and Behavior independently of each other in order to resolve the difficulties above.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.283-286
• /
• 2007

This paper proposes a genetic algorithm (GA) approach to instance selection in case-based reasoning (CBR) for the prediction of Korea Stock Price Index (KOSPI). CBR has been widely used in various areas because of its convenience and strength in complex problem solving. Nonetheless, compared to other machine learning techniques, CBR has been criticized because of its low prediction accuracy. Generally, in order to obtain successful results from CBR, effective retrieval of useful prior cases for the given problem is essential. However, designing a good matching and retrieval mechanism for CBR systems is still a controversial research issue. In this paper, the GA optimizes simultaneously feature weights and a selection task for relevant instances for achieving good matching and retrieval in a CBR system. This study applies the proposed model to stock market analysis. Experimental results show that the GA approach is a promising method for instance selection in CBR.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10b
• /
• pp.505-507
• /
• 1998

상속은 객체 지향 원리에서 만의 특성으로 추상화 레벨을 높여주고, 소프트웨어의 재사용을 강력히 지원하며, 대체 원리를 따른다. 또한 유지 보수의 용이성, 신뢰성등의 잇점을 제공한다. 그러나 측정 결과에 따르면 상속 계층이 깊어질수록 재사용이 어렵다고 조사되었으며 이는 재사용뿐아니라 개발에 있어서도 장애의 요인이 된다. 상속의 깊이를 최소화하기 위해서는 우선 상속 계층에서 직접적, 또는 간접적으로 사용되는 instance variable과 method만을 제외하고는 모두 삭제되어야 한다. 그러나, 다형성이 적용된 클래스는 정적(static) 분석이 불가능하므로 다형성을 제거하되, 다형성이 적용되었을 때와 동일한 모든 state, 기능 및 동작이 유지된 상태에서 처리되어야 한다. 다형성이 제거될 때 구현의 세부 사항은 변경하지 않음으로써 black box의 이점을 살린다. 다중상속의 경우는 각각의 상속 경로에 대하여 동일한 처리를 반복 수행하여 결과를 얻을 수 있으며, instance variable과 method의 access 레벨에 따라 처리 방법이 조금씩 달라진다. 본 논문에서는 C++에서의 다형성과 불필요한 instance variable과 method의 제거알고리즘에 대하여 논한다.

• Management Science and Financial Engineering
• /
• v.2 no.1
• /
• pp.147-176
• /
• 1996

The main objective of this study is to uncover the differences in the programming behavior between methodology creators and methodology users. We conducted an experiment with methodology creators who have invented one of the major object-oriented methodologies and with professional programmers who have used the same methodology for their software-development projects. In order to explain the difference between the two groups, we propose a theoretical framework that views programming as search in four problem spaces: representation, rule, instance and paradigm spaces. The main problem spaces in programming are the representation and rule spaces, while the paradigm and instance spaces are the supporting spaces. The results of the experiment showed that the methodology creators mostly adopted the paradigm space as their supporting space, while the methodology users chose the instance space as their supporting space. This difference in terms of the supporting space leads to different search behaviors in the main problem spaces, which in turn resulted in different final programs and performance.

• Bulletin of the Korean Mathematical Society
• /
• v.46 no.2
• /
• pp.245-253
• /
• 2009

A new hard problem called the vector decomposition problem (VDP) was recently proposed by Yoshida et al., and it was asserted that the VDP is at least as hard as the computational Diffie-Hellman problem (CDHP) under certain conditions. Kwon and Lee showed that the VDP can be solved in polynomial time in the length of the input for a certain basis even if it satisfies Yoshida's conditions. Extending our previous result, we provide the general condition of the weak instance for the VDP in this paper. However, when the VDP is practically used in cryptographic protocols, a basis of the vector space ${\nu}$ is randomly chosen and publicly known assuming that the VDP with respect to the given basis is hard for a random vector. Thus we suggest the type of strong bases on which the VDP can serve as an intractable problem in cryptographic protocols, and prove that the VDP with respect to such bases is difficult for any random vector in ${\nu}$.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2006.05a
• /
• pp.234-238
• /
• 2006

In this research, an analysis of damaged instance and forming fault for disc part in automotive steel wheel was performed. Rolled steel material, which had been used in the manufacturing of the damaged disc part, was prepared for tensile test, quantitative analysis of chemical component and acquirement of scanning electron microscope images. Although the results of mechanical properties and chemical component ratio for the material satisfied the suggested specification, some material inherent problem was found in the scanning electron microscope images. Finally, in an analysis of chemical component for the damaged disc part used in road condition, mismatching of chemical component ratio between the suggested specification and test result was found.

• Journal of Korea Multimedia Society
• /
• v.9 no.6
• /
• pp.762-772
• /
• 2006

Instance based learning algorithm is the best known lazy learner and has been successfully used in many areas such as pattern analysis, medical analysis, bioinformatics and internet applications. However, its feature weighting scheme is too naive that many other extensions are proposed. Our version of IB3 named as eXtended IBL (XIBL) improves feature weighting scheme by backward stepwise regression and its distance function by VDM family that avoids overestimating discrete valued attributes. Also, XIBL adopts leave-one-out as its noise filtering scheme. Experiments with common artificial domains show that XIBL is better than the original IBL in terms of accuracy and noise tolerance. XIBL is applied to two important applications - intrusion detection and spam mail filtering and the results are promising.

• Journal of the Korea Safety Management & Science
• /
• v.10 no.1
• /
• pp.117-128
• /
• 2008

The purpose of this paper is to cultivate field of new distribution and it will set three detail aims and such aims will be propelled. First, investigates a present condition of domestic financial distribution, second, presents standard model of financial distribution, and third, confirms suitability of standard model on financial distribution for instance analysis. incidental purpose is to offer improvement method about corresponding instance through deduction of problem and estimate of analysis instance. The purpose of this paper is also to induce more strategic financial industry through research of related system such as LIS, CVO and TMS or through research of prior innovation contents by application to NONGHYUP for realization of financial distribution model.