RISS 검색 - 학위논문 상세보기

다국어 초록 (Multilingual Abstract)

In this thesis, a radial basis function neural networks(RBFNNs) prediction model and pattern classifier designed with the aid of a novel learning method are proposed. The objective of this study is focused on the improvement of performance of the existing RBFNNs model by indroducing a new learning methods. In the case of the existing RBFNNs, outlier and noisy data included in dataset and the location of RBFs over the input space may be closely related to the performance. Based on these contents, two kinds of learning methods are proposed and brief explanation of the proposed learning method is enumerated as follows:
1) In the proposed prediction model and pattern classifier, the weighted FCM clustering is iteratively used for refinement of center point of each cluster. Clustering method is used to determine the center points over the input space by anaylzing the distribution of data patterns. The center points determined through the clustering method can be considered as the location of RBFs. In this study, conventional clustering method is used to initilize RBFs and then weighted FCM clustering driven with the aid of an auxiliary information is iteratively used for refinement of the locations of RBFs. Auxiliary information to be used of the weighted FCM can be obtained through sigmoid function and cross-entropy error function. Through this iterative refinement process of the location of RBFs, the performance of RBFNNs may be improved.
2) In order to trrain coefficients of connection weights between the hidden layer and output layer of the proposed prediction model and pattern classifier, margin-maximization is applied. Margin-maximization is a mechanism used to enhance the generalization ability of support vector machine(SVM) by maximizing distance from hyperplane to the nearest data pattern. Through the margin-maximization, weights each of data patterns can be obtained and then utilized to train the coefficients of connection weights. In this study, margin-maximization technique applied to least-square version of SVM(LS-SVM) is used. Since margin-maximization technique of original SVM should solve quadratic programming, lots of computational cost is consumed. Also, in regression problem, margin-maximization technique of general SVM cannot be applied. In contrast, margin-maximization technique of LS-SVM applies the least squares method to calculate weights instead of solving the quadratic programming, so computational cost is lower than original SVM. LS-SVM solves problem by implementing a linear equation, so it can apply to classification problem as well as regression problem.
From the viewpoint of performance improvement through the novel learning method, the proposed prediction model and pattern classifier are evaluated by using a variety of publicly available machine learning datasets and compared with a diverse of algorithms which realized to WEKA toolkt. In addition, the Friedman test is applied for statistical analysis of the proposed prediction model and pattern classifier. Furthermore, some practical application datasets such as the actifvated sludge process datasets, Portland cement datasets, plastic wastes datasets, and partial discharge datasets are also used to evaluate the performance.

번역하기

목차 (Table of Contents)

Ⅰ. 서 론 1
1. 연 구 목 적 1
1) 연구 가설 3
2) 연구의 독창성 4
2. 연 구 내 용 5

Ⅰ. 서 론 1
1. 연 구 목 적 1
1) 연구 가설 3
2) 연구의 독창성 4
2. 연 구 내 용 5
Ⅱ. 퍼지 클러스터링(FCM) 기반 방사형 기저 함수 신경회로망 8
1. 네트워크의 구조 8
2. 네트워크 학습을 위한 알고리즘 10
1) FCM을 사용한 방사형 기저 함수의 위치 학습 10
2) 최소 제곱 추정을 적용한 연결가중치의 계수 학습 12
Ⅲ. 가중 퍼지 클러스터링(Weighted FCM)을 사용한 방사형 기저 함수(RBF)의 위치 조정 14
1. 데이터 패턴별 가중치 계산 14
1) Sigmoid 함수를 사용한 가중치 계산 14
2) Cross-entropy 오차 함수를 사용한 가중치 계산 15
2. Weighted FCM을 사용한 RBF의 위치 조정 17
Ⅳ. 마진-최대화(Margin-Maximization) 기반 연결가중치의 계수 학습 20
1. 예측 모델에서의 연결가중치의 계수 학습 21
2. 패턴분류기에서의 연결가중치의 계수 학습 23
Ⅴ. Weighted FCM/Margin-Maximization을 사용한 방사형 기저 함수 신경회로망(RBFNN) 설계 27
1. Regression을 위한 RBFNN 예측 모델 28
2. Classification을 위한 RBFNN 패턴분류기 32
Ⅵ. 실험 및 결과 고찰 35
1. Regression을 위한 RBFNN 예측 모델 37
1) 평가 방법 37
2) 단일 입력기반 synthetic dataset 38
3) Publicly available regression dataset 44
4) 실적용 데이터(1): Portland cement 데이터 54
5) 실적용 데이터(2): 하수처리 데이터 57
2. Classification을 위한 RBFNN 패턴분류기 60
1) 평가 방법 60
2) Synthetic spiral datase 61
3) Publicly available classification dataset 64
4) 실적용 데이터(1): 폐플라스틱 데이터 72
5) 실적용 데이터(2): 부분방전 데이터 84
Ⅶ. 결 론 94
1. 결 론 94
2. 향후 연구과제 96
참 고 문 헌 97
ABSTRACT 109

참고문헌 (Reference)

1 S. Rosset , J. Zhu , T. Hastie ,, "Margin maximizing loss functions", vol . 16, 2003

2 A. K. Jain, "Data clustering : 50 years beyond K-means", vol . 31 , no . 8 , pp . 651-666, 2010

3 J. C. Bezdek , R. Ehrlich , W. Full, "FCM : The fuzzy c-means clustering algorithm", vol . 10 , no . 2-3 , pp . 191-203, 1984

4 T. C. Havens , J. C. Bezdek , C. Leckie , L. O . Hall , M. Palaniswami ,, "Fuzzy c-means algorithms for very large data", vol . 20 , no . 6 , pp . 1130-1146, 2012

5 D. E. Rumelhart , G. E. Hinton , and R. J. Williams, "Learning representations by back-propagating errors", vol . 323 , pp . 533-536, 1986

6 W. S. McCulloch , and W. H. Pitts ,, "A logical calculus of ideas immanent in nervous activity", vol . 5 , No . 4 , pp . 115-133 ,, 1943

7 P. D ’ Urso , P. Giordani ,, "A weighted fuzzy c-means clustering model for fuzzy data", vol . 50 , no . 6 , pp . 1496-1523, 2006

8 E.-H. Kim , S.-K. Oh , W. Pedrycz, "Reinforced rule-based fuzzy models : Design and analysis", vol . 119 , pp . 44-58, 2017

9 S.-K. Oh , W. Pedrycz , B.-J . Park, "Polynomial neural networks architecture : analysis and design", vol . 29 , no . 6 , pp . 703-725, 2003

10 T. M. Kodinariya , P. R. Makwana ,, "Review on determining number of Cluster in K-Means Clustering", vol . 1 , no . 6 , pp . 90-95, 2013