• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.241-243
• /
• 2004

실제 응용에서 제기되는 많은 최적화 문제는 실제로 여러 개의 목적함수를 가진 최적화 문제로 분류될 수 있다. 이러한 다중 목적함수 최적화 문제에 적용되온 방법 중에서 다중 목적함수 진화 알고리즘은 해집합을 이용한다는 특성 및 목적함수 처리의 용이성 때문에 많은 연구가 이루어지고 있다. 본 논문에서는 대표적인 다중 목적함수 진화 알고리즘이라 할 수 있는 입실론-다중 목적함수에 대하여 다양한 최적화 문제에 대하여 실험적으로 비교 분석해 보았다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.87-95
• /
• 1996

A new design scheme for the fuzzy pattern matching classifier (FPMC) is proposed. in conventional design of FPMC, there are no exact information about the membership function of which shape and number critically affect the performance of classifier. So far, a trial and error or heuristic method is used to find membership functions for the input patterns. But each of them have limits in its application to the various types of pattern recognition problem. In this paper, a new method to find the appropriate shape and number of membership functions for the input patterns which minimize classification error is proposed using genetic algorithms(GAs). Genetic algorithms belong to a class of stochastic algorithms based on biological models of evolution. They have been applied to many function optimization problems and shown to find optimal or near optimal solutions. In this paper, GAs are used to find the appropriate shape and number of membership functions based on fitness function which is inversely proportional to classification error. The strings in GAs determine the membership functions and recognition results using these membership functions affect reproduction of next generation in GAs. The proposed design scheme is applied to the several patterns such as tire tread patterns and handwritten alphabetic characters. Experimental results show the usefulness of the proposed scheme.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1343-1344
• /
• 2015

본 연구에서는 증분형 FCM(Incremental Fuzzy C-Means: Incremental FCM) 클러스터링 알고리즘을 기반으로 방사형 기저함수 신경회로망(Radial Basis Function Neural Networks: RBFNN) 패턴 분류기를 설계한다. 방사형 기저함수 신경회로망은 조건부에서 가우시안 함수 또는 FCM을 사용하여 적합도를 구하였지만, 제안된 분류기에서는 빅 데이터간의 적합도를 구하기 위해 증분형 FCM을 사용한다. 또한, 빅 데이터를 학습하기 위해 결론부에서 재귀최소자승법(Recursive Least Square Estimation: RLSE)을 사용하여 다항식 계수를 추정한다. 마지막으로 추론부에서는 증분형 FCM에서 구한 적합도와 재귀최소자승법으로 구한 다항식을 이용하여 최종 출력을 구한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.655-660
• /
• 2003

In general, Support Vector Machine has a good performance in binary classification, but it has the limitation on multi-pattern classification. So, we proposed an Import Vector Voting model for two or more labels classification. This model applied kernel bagging strategy to Import Vector Machine by Zhu. The proposed model used a voting strategy which averaged optimal kernel function from many kernel functions. In experiments, not only binary but multi-pattern classification problems, our proposed Import Vector Voting model showed good performance for given machine learning data.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.321-324
• /
• 2004

본 논문에서는 유도전동기 고장진단을 위하여 계층적인 하이브리드 뉴럴네트웍을 제안하였다. 시스템의 입출력 데이터에 근거하여 패턴을 분류하고자 할 때 직접적인 분류가 어렵거나 성능이 좋지 않을 경우 적절한 방법을 통하여 변환을 하거나 또는 패턴 분류기의 특성에 맞도록 변환하여 패턴 분류 성능을 향상하는 등 단계별 변환 및 분류 기법을 이용하였다. 제안된 방법에서는 실험에 의해 측정된 전류값을 주기별로 주성분분석(PCA) 기법을 이용하여 입력차원을 축소한 후 이를 조건부 FCM으로 방사기저함수의 초기치를 최적화하여 학습을 하였다. 이는 주성분분석이 가지는 특성을 이용하여 데이터의 특징을 나누었으며 이를 뉴럴네트웍의 학습 기능을 이용하여 모델의 최종 성능을 개선하는 것이다. 각각의 알고리즘이 가지는 특징을 활용하면서도 단점을 계층적으로 보안하여 유도 전동기 고장 진단 성능을 개선하였다. 이를 실제 계측된 유도전동기 데이터를 이용하여 제안된 방법의 유용성을 보이고자 한다.

• Journal of KIISE:Databases
• /
• v.28 no.4
• /
• pp.545-557
• /
• 2001

In this paper we propose a hierarchical classification algorithm based on rough membership function which can reason a new object approximately. We use the fuzzy reasoning method that substitutes fuzzy membership value for linguistic uncertainty and reason approximately based on the composition of membership values of conditional sttributes Here we use the rough membership function instead of the fuzzy membership function It can reduce the process that the fuzzy algorithm using fuzzy membership function produces fuzzy rules In addition, we transform the information system to the understandable minimal decision information system In order to do we, study the discretization of continuous valued attributes and propose the discretization algorithm based on the rough membership function and the entropy of the information theory The test shows a good partition that produce the smaller decision system We experimented the IRIS data etc. using our proposed algorithm The experimental results with IRIS data shows 96%~98% rate of classification.

• Korean Journal of Remote Sensing
• /
• v.20 no.6
• /
• pp.419-425
• /
• 2004

This study has proposed a data fusion method based on the Dempster-Shafer evidence theory The Dempster-Shafer fusion uses mass functions obtained under the assumption of class-independent Gaussian assumption. In the Dempster-Shafer approach, uncertainty is represented by 'belief interval' equal to the difference between the values of 'belief' function and 'plausibility' function which measure imprecision and uncertainty By utilizing the Dempster-Shafer scheme to fuse the data from multiple sensors, the results of classification can be improved. It can make the users consider the regions with mixed classes in a training process. In most practices, it is hard to find the regions with a pure class. In this study, the proposed method has applied to the KOMPSAT-EOC panchromatic image and LANDSAT ETM+ NDVI data acquired over Yongin/Nuengpyung. area of Kyunggi-do. The results show that it has potential of effective data fusion for multiple sensor imagery.

• The Korean Journal of Applied Statistics
• /
• v.35 no.5
• /
• pp.593-601
• /
• 2022

We call "fruad" the cases that are not frequently occurring but cause significant losses. Fraud detection is commonly encountered in various applications, including wafer production in the semiconductor industry. It is not trivial to directly extend the standard binary classification methods to the fraud detection context because the misclassification cost is much higher than the normal class. In this article, we propose the functional fraud detection support vector machine (F²DSVM) that extends the fraud detection support vector machine (FDSVM) to handle functional covariates. The proposed method seeks a classifier for a function predictor that achieves optimal performance while achieving the desired sensitivity level. F²DSVM, like the conventional SVM, has piece-wise linear solution paths, allowing us to develop an efficient algorithm to recover entire solution paths, resulting in significantly improved computational efficiency. Finally, we apply the proposed F²DSVM to the defective wafer detection problem and assess its potential applicability.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.4
• /
• pp.700-713
• /
• 2009

Most of any existing classification algorithm in data mining area have focused on goals improving efficiency, which is to generate decision tree more rapidly by utilizing just less computing resources. In this paper, we focused on the efficiency as well as effectiveness that is able to generate more meaningful classification rules in application area, which might consist of the ontology automatic generation, business environment and so on. For this, we proposed not only novel function with the weight of support and confidence degree but also analyzed the characteristics of the weighted function in theoretical viewpoint. Furthermore, we proposed novel classification algorithm based on the weighted function and the characteristics. In the result of evaluating the proposed algorithm, we could perceive that the novel algorithm generates more classification rules with significance more rapidly.

• Convergence Security Journal
• /
• v.20 no.4
• /
• pp.81-90
• /
• 2020

In the real world, new types of attacks or variants are constantly emerging, but attack traffic classification models developed through artificial neural networks and supervised learning do not properly detect new types of attacks that have not been trained. Most of the previous studies overlooked this problem and focused only on improving the structure of their artificial neural networks. As a result, a number of new attacks were frequently classified as normal traffic, and attack traffic classification performance was severly degraded. On the other hand, the softmax function, which outputs the probability that each class is correctly classified in the multi-class classification as a result, also has a significant impact on the classification performance because it fails to calculate the softmax score properly for a new type of attack traffic that has not been trained. In this paper, based on this characteristic of softmax function, we propose an efficient method to improve the classification performance against new types of attacks by classifying traffic with a probability below a certain level as attacks, and demonstrate the efficiency of our approach through experiments.