• International Journal of Control, Automation, and Systems
• /
• 제2권1호
• /
• pp.118-124
• /
• 2004

This paper proposes an approach to deriving a fuzzy classifier based on evolutionary supervised clustering, which identifies the optimal clusters necessary to classify classes. The clusters are formed by multi-dimensional weighted Euclidean distance, which allows clusters of varying shapes and sizes. A cluster induces a Gaussian fuzzy antecedent set with unique variance in each dimension, which reflects the tightness of the cluster. The fuzzy classifier is com-posed of as many classification rules as classes. The clusters identified for each class constitute fuzzy sets, which are joined by an "and" connective in the antecedent part of the corresponding rule. The approach is evaluated using six data sets. The comparative results with different classifiers are given.are given.

• 전기학회논문지
• /
• 제66권7호
• /
• pp.1131-1140
• /
• 2017

This study is concerned with a design methodology of optimized fuzzy clustering-based neural network classifier for classifying black plastic. Since the amount of waste plastic is increased every year, the technique for recycling waste plastic is getting more attention. The proposed classifier is on a basis of architecture of radial basis function neural network. The hidden layer of the proposed classifier is composed to FCM clustering instead of activation functions, while connection weights are formed as the linear functions and their coefficients are estimated by the local least squares estimator (LLSE)-based learning. Because the raw dataset collected from Raman spectroscopy include high-dimensional variables over about three thousands, principal component analysis(PCA) is applied for the dimensional reduction. In addition, artificial bee colony(ABC), which is one of the evolutionary algorithm, is used in order to identify the architecture and parameters of the proposed network. In experiment, the proposed classifier sorts the three kinds of plastics which is the most largely discharged in the real world. The effectiveness of the proposed classifier is proved through a comparison of performance between dataset obtained from chemical analysis and entire dataset extracted directly from Raman spectroscopy.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 춘계학술대회 학술발표 논문집
• /
• pp.187-190
• /
• 2001

Due to the recent proliferation of the internet, it is broadly granted that the necessity of the automatic document categorization has been on the rise. Since it is a heavy time-consuming work and takes too much manpower to process and classify manually, we need a system that categorizes them automatically as their contents. In this paper, we propose the automatic E-mail response system that is based on 2 hierarchical document clustering methods. One is to get the final result from the classifier trained seperatly within each class, after clustering the whole documents into 3 groups so that the first classifier categorize the input documents as the corresponding group. The other method is that the system classifies the most distinct classes first as their similarity, successively. Neural networks have been adopted as classifiers, we have used dendrograms to show the hierarchical aspect of similarities between classes. The comparison among the performances of hierarchical and non-hierarchical classifiers tells us clustering methods have provided the classification efficiency.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권9호
• /
• pp.857-868
• /
• 2007

좋은 분류기 앙상블은 분류기간에 상호 보완성을 갖추어 높은 인식 성능을 보여야 하며, 크기가 작아 계산 효율이 좋아야 한다. 이 논문은 이러한 목적을 달성하기 위한 거친-섬세한 (coarse-to-fine)단계를 밟는 분류기 앙상블 선택 방법을 제안한다. 이 방법이 성공하기 위해서는 초기 분류기 풀 (pool)이 충분히 다양해야 한다. 이 논문에서는 여러 개의 서로 다른 분류 알고리즘과 아주 많은 수의 특징 부분집합을 결합하여 충분히 큰 분류기 풀을 생성한다. 거친 선택 단계에서는 분류기 풀의 크기를 적절하게 줄이는 것이 목적이다. 분류기 군집화 알고리즘을 사용하여 다양성을 최소로 희생하는 조건하에 분류기 풀의 크기를 줄인다. 섬세한 선택에서는 유전 알고리즘을 이용하여 최적의 앙상블을 찾는다. 또한 탐색 성능이 개선된 혼합 유전 알고리즘을 제안한다. 널리 사용되는 필기 숫자 데이타베이스를 이용하여 기존의 단일 단계 방법과 제안한 두 단계 방법의 성능을 비교한 결과 제안한 알고리즘이 우수함을 입증하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.117-118
• /
• 2007

A robot needs a human detection algorithm for interaction with a human. This paper proposes a method that finds people using a SVM (support vector machine) classifier and a stereo camera. Feature vectors of SVM are extracted by HoG (histogram of gradient) within images. After training extracted vectors from the clustered images, the SVM algorithm creates a classifier for human detection. Each candidate for a human in the image is generated by clustering of depth information from a stereo camera and the candidate is evaluated by the classifier. When compared with the existing method of creating candidates for a human, clustering reduces computational time. The experimental results demonstrate that the proposed approach can be executed in real time.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.2037-2040
• /
• 2004

In this paper we introduce a new prototype based classifiers for overlapping data, where training pattern can be overlap on the feature space. The proposed classifier is based on the prototype from neural network classifier (NNC)[1] for overlap data. The method automatically chooses the initial center and two radiuses for each class. The center is used as a mean representative of training data for each class. The unclassified pattern is classified by measure distance from the class center. If the distance is in the lower (shorter radius) the unknown pattern has the high percentage of being in this class. If the distance is between the lower and upper (further radius), the pattern has the probability of being in this class or others. But if the distance is outside the upper, the pattern is not in this class. We borrow the words upper and lower from the rough set to represent the region of certainty [3]. The training algorithm to find number of cluster and their parameters (center, lower, upper) is presented. The clustering result is tested using patterns from Thai handwritten letter and the clustering result is very similar to human eyes clustering.

• 대한전기학회논문지:전력기술부문A
• /
• 제50권3호
• /
• pp.112-119
• /
• 2001

This paper proposes an adaptive reclosing scheme which is based on the classification of fault patterns. In case that the first reclosing is unsuccessful in distribution system employing with two-shot reclosing scheme, the proposed method can determine whether the second reclosing will be attempted of not. If the first reclosing is unsuccessful two fault currents can be measured before the second reclosing is attempted, where these two fault currents are utilized for an adaptive reclosing scheme. Total harmonic distortion and RMS are used for extracting the characteristics of two fault currents. And the pattern of two fault currents is respectively classified using a mountain clustering method a minimum-distance classifier. Mountain clustering method searches the cluster centers using the acquired past data. And minimum-distance classifier is used for classifying the measured two currents into one of the searched centers respectively. If two currents have the different pattern it is interpreted as temporary fault. But in case of the same pattern, the occurred fault is interpreted as permanent. The proposed method was tested for the fault data which had been measured in KEPCO's distribution system, and the test results can demonstrate the effectiveness of the adaptive reclosing scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제3권2호
• /
• pp.134-146
• /
• 2009

Traffic classification seeks to assign packet flows to an appropriate quality of service(QoS) class based on flow statistics without the need to examine packet payloads. Classification proceeds in two steps. Classification rules are first built by analyzing traffic traces, and then the classification rules are evaluated using test data. In this paper, we use self-organizing map and K-means clustering as unsupervised machine learning methods to identify the inherent classes in traffic traces. Three clusters were discovered, corresponding to transactional, bulk data transfer, and interactive applications. The K-nearest neighbor classifier was found to be highly accurate for the traffic data and significantly better compared to a minimum mean distance classifier.

• 정보처리학회논문지B
• /
• 제11B권4호
• /
• pp.485-490
• /
• 2004

자조의 표현에서 군집화는 주어진 데이터를 서로 유사한 개체들끼리 몇 개의 집단으로 묶는 작업을 수행한다. 군집화의 유사도 결정 측도는 맡은 연구들에서 매우 다양한 것들이 사용되었다. 하지만 군집화 결과의 성능 측정에 대한 객관적인 기준 설정이 어렵기 때문에 군집화 결과에 대한 해석은 매우 주관적이고, 애매한 경우가 많다. 퍼지 군집화는 이러한 주관적인 군집화 문제에 있어서 객관성 있는 군집 결정 방안을 제시하여 준다. 각 개체들이 특정 군집에 속하게 될 퍼지 멤버 함수값을 원소로 하는 유사도 행렬을 통하여 군집화를 수행한다. 본 논문에서는 차원 축소기법의 하나인 주성분 분석과 강력한 통계적 학습 이론인 베이지안 학습을 결합한 군집화 모형을 제안하여, 객관적인 퍼지 군집화를 수행하였다. 제안 알고리즘의 성능 평가를 위하여 UCI Machine Loaming Repository의 Iris와 Glass Identification 데이터를 이용한 실험 결과를 제시하였다.

• 한국지능시스템학회논문지
• /
• 제22권1호
• /
• pp.81-86
• /
• 2012

본 논문에서는 전형적인 Linear Discriminant Analysis을 확장시켜 전체 입력공간을 다수의 지역공간으로 분할하고 분할된 공간에 Local Linear Discriminant Analysis 기반으로 하여 패턴 분류 규칙을 설계하는 새로운 방법을 제안한다. 전체 입력공간을 여러 개의 지역공간으로 분할하기 위한 방법으로 unsupervised clustering의 대표적인 방법인 k-Means 클러스터링 기법과 최적화 알고리즘인 차분 진화 연산 알고리즘을 사용한다. 제안된 알고리즘의 성능 평가를 위해 기존의 패턴 분류기와 비교 결과를 제시한다.