• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2846-2849
• /
• 2002

본 논문은 뉴로-퍼지 시스템에서의 규칙 선택 및 모델 학술에 대하여 EM 알고리즘을 기반으로 하는 구조 동정을 제안한다. 뉴로-퍼지 모델링에서의 초기 파라미터가 학습과정에서의 모델 성능에 큰 영향을 주고 있다. 주어진 데이터에 근거한 파라미터 추정에는 다양한 방법들이 소개되고 응용되어져 왔는데 이전 연구들에서 볼 수 있는 HCM, FCM 등은 데이터와의 유클리디언 거리를 최소화하는 중심점을 파라미터로 선택하는 등의 방법과 퍼지 균등화 등은 데이터의 확률 밀도함수를 이용하여 파라미터를 추정하였다. 제안된 방법에서는 데이터에서의 Maximum Likelihood Estimator를 기반으로 하는 방법으로 EM 알고리즘을 이용하였다. 초기 파라미터의 결정에서 EM 알고리즘을 이용하여 뉴로-퍼지 모델의 전제부 소속함수 파라미터 추정을 실시한다. EM 알고리즘을 이용한 퍼지 모델의 특징으로는 전제부가 클러스터링에 의하여 생성되므로 입력의 차원이나 소속함수의 수가 증가하여도 규칙의 수는 증가하지 않는다. 이를 자동차 MPG 예제를 통하여 제안된 방법의 유용성을 보이고자 한다.

• Journal of Korean Society of Transportation
• /
• v.29 no.1
• /
• pp.125-134
• /
• 2011

In order to reconstruct vehicle-pedestrian collision accidents, this paper presents a fuzzy tool to estimate accurately the impact velocity of the vehicle using parameters which could be easily collectable at the accident scene. The fuzzy rules and membership functions were set up using number of over 200 domestic and foreign data from accidents and empirical tests and 700 data from multibody simulation experiments. The developed fuzzy tool deduces the category of pedestrian trajectory and impact speed of the vehicle using 4 membership functions and 2 logic rules. The membership function of throw distance was differently set according to the deduced category of trajectories. The implemented fuzzy program was validated through comparing with the domestic and foreign empirical data. The output results agree very well in impact velocities of vehicle resulting the accuracy and usefulness of the developed tool in the reconstruction analysis of vehicle-pedestrian collision accidents.

• Journal of KIISE:Software and Applications
• /
• v.32 no.1
• /
• pp.1-9
• /
• 2005

In this paper we propose a modified fuzzy min-max neural network model for pattern classification and discuss the usefulness of the model. We define a new hypercube membership function which has a weight factor to each of the feature within a hyperbox. The weight factor makes it possible to consider the degree of relevance of each feature to a class during the classification process. Based on the proposed model, a knowledge extraction method is presented. In this method, a list of relevant features for a given class is extracted from the trained network using the hyperbox membership functions and connection weights. Ft)r this purpose we define a Relevance Factor that represents a degree of relevance of a feature to the given class and a similarity measure between fuzzy membership functions of the hyperboxes. Experimental results for the proposed methods and discussions are presented for the evaluation of the effectiveness and feasibility of the proposed methods.

• Economic and Environmental Geology
• /
• v.50 no.3
• /
• pp.195-214
• /
• 2017

Statistical landslide susceptibility analysis, which is widely used among various landslide susceptibility analysis approaches, predicts the unstable area by analyzing statistical relationship between landslide occurrence locations and landslide controlling factors. However, uncertainties are involved in the procedures of the susceptibility analysis and therefore, fuzzy approach has been used to deal properly with uncertainties. The fuzzy approach used fuzzy set theory and fuzzy membership function to quantify uncertainties involved in landslide controlling factors. Various fuzzy approaches were suggested in the procedure of the membership value determination and fuzzy operation in the previous researches. However, few studies were carried out to compare the analysis results obtained from various approaches for membership function determination and fuzzy operation. Therefore, in this study, the authors selected Jinbu area, which a large number of landslides were occurred at in 2006, to apply two most commonly used methods, the frequency ratio and the cosine amplitude method to derive membership values for each controlling factor. In addition, the integration of different thematic layers to produce landslide susceptibility map was performed by several fuzzy operators such as AND, OR, algebraic product, algebraic sum and Gamma operator. The results of the landslide susceptibility analysis using two different methods for the determination of fuzzy membership values and various fuzzy operators were compared on the basis of ROC graph to check the feasibility of the fuzzy based landslide susceptibility analysis.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.1
• /
• pp.32-40
• /
• 2008

Fuzzy rules are very useful and efficient to describe classification rules especially when the attribute values are continuous and fuzzy in nature. However, it is generally difficult to determine membership functions for generating efficient fuzzy classification rules. In this paper, we propose a method of automatic generation of efficient fuzzy classification rules using evolutionary algorithm. In our method we generate a set of initial membership functions for evolutionary algorithm by supervised clustering the training data set and we evolve the set of initial membership functions in order to generate fuzzy classification rules taking into consideration both classification accuracy and rule comprehensibility. To reduce time to evaluate an individual we also propose an evolutionary algorithm with data partition evaluation in which the training data set is partitioned into a number of subsets and individuals are evaluated using a randomly selected subset of data at a time instead of the whole training data set. We experimented our algorithm with the UCI learning data sets, the experiment results showed that our method was more efficient at average compared with the existing algorithms. For the evolutionary algorithm with data partition evaluation, we experimented with our method over the intrusion detection data of KDD'99 Cup, and confirmed that evaluation time was reduced by about 70%. Compared with the KDD'99 Cup winner, the accuracy was increased by 1.54% while the cost was reduced by 20.8%.

• 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.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.599-604
• /
• 2005

This paper presents fuzzy rules to predict diagnosis of Wisconsin breast cancer with minimized number of feature in put using the neural network with weighted fuzzy membership functions (NEWFM) and the non-overlap area distribution measurement method. NEWFM is capable of self-adapting weighted membership functions from the given the Wisconsin breast cancer clinical training data. n set of small, medium, and large weighted triangular membership functions in a hyperbox are used for representing n set of featured input. The membership functions are randomly distributed and weighted initially, and then their positions and weights are adjusted during learning. After learning, prediction rules are extracted directly from n set of enhanced bounded sums of n set of small, medium, and large weighted fuzzy membership functions. Then, the non-overlap area distribution measurement method is applied to select important features by deleting less important features. Two sets of prediction rules extracted from NEWFM using the selected 4 input features out of 9 features outperform to the current published results in number of set of rules, number of input features, and accuracy with 99.71%.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.27-32
• /
• 2012

In this paper, we propose a fuzzy method that improves the existing problem of the fuzzy filtering algorithm. The proposed fuzzy filtering algorithm separates R, G, and B channels from the color image. Mask information was extracted from separated channels and the brightness of the mean value and median value for channels was applied in the function of the proposed fuzzy method to calculate the membership and achieve application in the inference rule. Also, the membership degrees of R, G, and B were used to distinguish the possibility of noise. The proposed fuzzy method selected three membership functions. If noise is distinguished, the noise is eliminated by selecting the median value or mean value as the relevant pixel value according to the degree of noise. By applying the proposed method in color images, it was verified that the proposed method is more effective in eliminating noise when compared with the conventional fuzzy filtering method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.5
• /
• pp.1-12
• /
• 2000

In this paper we propose a design method of the adaptive fuzzy controller for autonomous navigation of mobile robots based on the fuzzy theory. We present two improvements. First, unnecessary rules in the fuzzy inference process make data processing time increase. We reduce this data processing time by generating suitable fuzzy inference rules and membership functions according to the current state of a mobile robot. It is implemented with the clustering method using input and output data pairs, and then it is possible for a mobile robot to navigate in shorter processing time with less fuzzy inference rules. Second, existing algorithms used fixed membership functions of input and output variables, hence converged slowly. We improve convergence time via scaling membership functions generated by the clustering method. To evaluate and compare the performance of the proposed method with the existing fuzzy navigation controller, computer simulations and navigation experiments of a mobile robot are Presented.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.10
• /
• pp.2485-2492
• /
• 1997

Nowadays the traffic conditions have been getting worse due to continuous increase in the number of vehicles. So it has become more important to manage traffic signal lights efficiently. Recently fuzzy logic is introduced to control the cycle time of traffic lights adaptively. Conventional fuzzy logic controller adjusts the extension time of current green phase by using the fuzzy input variables such as the number of entering vehicles at the green light and the number of waiting vehicle during the red light. However this scheme is inadequate for an intersection with variable traffic densities. In this paper, a new FLC with asymmetric membership functions that reflects more exactly traffic flows than other FLCs with symmetric ones regardless of few control rules is propsed. The effectiveness of the proposed method was shown through simulation of a single intersection. The experimental results yielded the superior performance of the proposed FLC in terms of the average delay time, the number of passed vehicles, and the degree of saturation.