-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
763-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Noboru YAMAGUCHI
Article type: Article
1998 Volume 10 Issue 5 Pages
764-774
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Atsushi SHIMOJIMA
Article type: Article
1998 Volume 10 Issue 5 Pages
775-784
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Suguru YAMAGUCHI
Article type: Article
1998 Volume 10 Issue 5 Pages
785-795
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Ken-ichi MATSUMOTO
Article type: Article
1998 Volume 10 Issue 5 Pages
796-803
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Yoshio SHIMIZU
Article type: Article
1998 Volume 10 Issue 5 Pages
804-812
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Mitsuhiro SATO
Article type: Article
1998 Volume 10 Issue 5 Pages
813-826
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Taiji YAMADA
Article type: Article
1998 Volume 10 Issue 5 Pages
827-835
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Article type: Bibliography
1998 Volume 10 Issue 5 Pages
836-841
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Hideyuki TAKAGI
Article type: Article
1998 Volume 10 Issue 5 Pages
842-843
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Hiroshi KUTSUMI, Jun OZAWA, Kouji MIURA, Takeshi IMANAKA, Atsuo TSUJI, ...
Article type: Article
1998 Volume 10 Issue 5 Pages
844-847
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Kayo IMAMURA
Article type: Article
1998 Volume 10 Issue 5 Pages
848-849
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Yasuhiro SUZUKI, Hiroshi TANAKA
Article type: Article
1998 Volume 10 Issue 5 Pages
850-852
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Tomohiro YOSHIKAWA
Article type: Article
1998 Volume 10 Issue 5 Pages
853-858
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Makoto FUJIYOSHI
Article type: Article
1998 Volume 10 Issue 5 Pages
859-861
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
862-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
863-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
863-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
864-865
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
866-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
[in Japanese]
Article type: Article
1998 Volume 10 Issue 5 Pages
866-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
Minoru TADA, Hiroaki ISHII
Article type: Article
1998 Volume 10 Issue 5 Pages
867-875
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
The objective of the ordinary assignment problem is maximizing the total efficiency when each worker is assigned to exactly one job and each job receives exactly one worker. In some situation, however, it is not so easy to determine every efficiency rigidly. In this paper, we accordingly consider the assignment problem as a "decision making in fuzzy environment". That is, the weights of jobs are presented by membership functions and introduce fuzzy efficiencies characterized by fuzzy numbers. In this case, we can formulate and solve the problem by making use of the agreement index. Further, in order to consider more flexible situation, we characterize each worker's satisfaction degree for each job by the corresponding membership functions and formulate the problem with two objectives, i. e., (1)the total efficiency to be maximized and (2)the minimum among all satisfaction degrees to be maximized. In other words, the bi-criteria problem is formulated from the viewpoint of not only (1) DM (Decision Maker) but also (2) workers. To solve the bi-criteria problem, we propose an efficient algorithm which is the interactive system with the concept of non-dominated solution.
View full abstract
-
Masataka TOKUMARU, Kazumi YAMASHITA, Noriaki MURANAKA, Shigeru IMANISH ...
Article type: Article
1998 Volume 10 Issue 5 Pages
876-887
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
The purpose of this paper is to propose a user interface using the expert system. This user interface is loaded in the interactive harmonization system. The user interface reflects the ambiguousness contained in the system and the user's subjectivity and taste in the technical work. Each membership function, which is defined by the control parameter specified by the user and based on seven pieces of technical musical knowledge, is used for this harmonization system. By changing the value of the control parameter according to the experimental music and examining the result, we find another natural accompaniment chord.
View full abstract
-
Yinzhen LI, Mitsuo GEN
Article type: Article
1998 Volume 10 Issue 5 Pages
888-898
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
In this paper, we present a new approach which uses spanning tree-based genetic algorithm for solving bicriteria transportation problem. The transportation problem has the special data structure in solution characterized as a transportation graph. In encoding transportation problem, we absorb the concept on spanning tree and adopt the Prufer number as it is capable of representing all possible basic solutions. The crossover and mutation were designed based on this encoding. And we designed the criterion by which chromosomes can be always feasibly converted to a transportation tree. In the evolutionary process, the mixed strategy with (μ+λ)-selection and roulette wheel selection is used. To confirm the performance of the proposed algorithm based on spanning tree representation, we firstly show the effectiveness of the algorithm by the single objective problems. And then the efficiency of the proposed algorithm will shown.
View full abstract
-
Katsuari KAMEI, Takahito FUKUOKA
Article type: Article
1998 Volume 10 Issue 5 Pages
899-906
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
Learning Vector Quantization (LVQ) proposed by T.Kohonen is a simple algorithm and also has a strong pattern recognition power. It is possible for Fuzzy Learning Vector Quantization (FLVQ) to construct networks with a higher ability to extract characteristics of pattern data than that of LVQ. This paper proposes a pattern matching method based on the learning algorithm of FLVQ for handwritten Japanese character (KANJI) recognition. The authors make it clear through recognition experiments that the proposed method is useful even for complex structured characters but still has some difficulties.
View full abstract
-
Kazutaka UMAYAHARA, Sadaaki MIYAMOTO
Article type: Article
1998 Volume 10 Issue 5 Pages
907-914
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
A new method of fuzzy c-means using a quadratic term of the membership metrix components as a regularizing function is proposed. This method is accordingly called a quadratic regularization. It should be noted that the standard fuzzy c-means can be ragarded as a regularization of the crisp k-means, and moreover another regularization method by using an entropy term has already been proposed. An algorithm for calculating membership values is derived, whereas the formula for cluster centers is similar to the standard method of fuzzy c-means. The present method of quadratic regularization yields a piecewise linear function of fuzzy classification. We have thus the third method of fuzzy c-means in addition to the standard method and the entropy method previously proposed.
View full abstract
-
Michio SUGENO, Chang-Hoon LEE
Article type: Article
1998 Volume 10 Issue 5 Pages
915-936
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
This paper is concerned with the design of fuzzy controllers based on the fuzzy models of plants. The models are the simplified ones used in conventional fuzzy control. To begin with, we introduce canonical forms of unforced fuzzy systems with singleton consequents and a stability theorem given by one of the authors. Next, we improve the theorem and suggest a stabilization algorithm for fuzzy control systems. The basic method is, as we usually do in control theory, to find a fuzzy controller so that a closed system becomes stable, where we use an inverse-problem-approach of optimal control for finding a stabilizing control. The stability analysis and the design algorithm are verified by their applications to nonlinear systems.
View full abstract
-
Shoji HIRANO, Naotake KAMIURA, Yutaka HATA
Article type: Article
1998 Volume 10 Issue 5 Pages
937-946
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
In this paper, we propose a segmentation method of the brain portions based on fuzzy inference mechanism. Our method employs knowledge of location, knowledge of boundary surface and knowledge of intensity to segment the whole brain region into the left cerebral hemisphere, the right cerebral hemisphere, the cerebellum and the brain stem. Each of their knowledge can provide a degree for the boundary on their fuzzy if-then rules. The segmentation procedure has two steps. First, the location knowledge roughly segments whole brain into four clusters which have primitive feature about their location. Second, other two knowledge infer the possibility to the boundary among their portions. Their inference results are combined into a total degree, which is evaluated to determine the precise boundary. We applied our method to 36 human brain MR data. The average error ratio against medical doctor's manual segmentation was 2.5%.
View full abstract
-
Akio OHMORI, Takehisa ONISAWA
Article type: Article
1998 Volume 10 Issue 5 Pages
947-956
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
In this paper an analysis model of human reliability is presented, which employs natural language and a fault tree. The meaning of linguistic term about human reliability is expressed by a subjective measure of unreliability defined on subjective unreliability [0, 1], and its AND operation and OR operation in a fault tree analysis are defined. Analysis results are expressed by natural language. In this paper a dependency among human tasks and a task procedure are especially considered as important points in a human reliability analysis. There are two kinds of dependencies among consecutive similar tasks that are failure dependency and success dependency. The latter is peculiar to human reliability. The task procedure is often prescribed in tasks. In this case the error in the task procedure must be considered in the analysis. This is also peculiar to human reliability. Finally, human reliability in tasks considering the dependency and the task procedure is analyzed as an example and analysis results are discussed in order to confirm the validity of the presented analysis model.
View full abstract
-
Eiji NAKANISHI, Yoshihiro MORI, Takehiro MORI, Yasuaki KUROE
Article type: Article
1998 Volume 10 Issue 5 Pages
957-967
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
Fuzzy control systems can be regarded as nonlinear systems due to the input-output relations of the controller. Hence, various stability analysis methods for nonlinear systems have been successfully applied to fuzzy control systems. But such studies investigate certain specific fuzzy control systems only to get ad hoc results. A possible strategy to cope with such a situation is to increase the number of usable tools so as to cover a wide range of systems. This prompts us to add an alternative to the possible analysis tools. In this paper, we explore a stability test method by Haddad for fuzzy control systems. This method has the merit in that it can be used for many types of nonlinear systems, because it does not need any constraints to the nonlinear property such as sector conditions. The method boils down to a nonlinear optimization problem, which is hard to solve. Using the local monotonicity property of input-output relation of the controller employed here, however, we resolve this problem to obtain an effective stability test. Numerical examples are provided to show comparison with the circle condition and simulation results.
View full abstract
-
Osamu MORIKAWA
Article type: Article
1998 Volume 10 Issue 5 Pages
968-972
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
We introduce the family of fuzzy temporal operations. This paper proposes the finite fuzzy modal logic with fuzzy temporal operations. After giving syntax and semantics, we present a formal system in an extended Gentzen style. Then we prove the soundness theorem as well as the completeness theorem.
View full abstract
-
Daouren F. AKHMETOV, Yasuhiko DOTE, Yukinori SUZUKI, Sato SAGA
Article type: Article
1998 Volume 10 Issue 5 Pages
973-982
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
In this paper an automatic fuzzy rule generation problem through the artificial neural network (ANN) approach is considered. The unknown fuzzy relation reconstruction problem is treated as an optimization of the structure and parameters of the neural network. The functional equivalence between some classes of fuzzy systems and radial basis function networks (RBFNs), namely, their localized sensitivity to input value, is a background of the proposed approach. RBFN with improved structure and advanced learning feature is developed based on General Parameter (GP) method of complex system identification. Main characteristics of the GP method are independence of the learning speed from the dimensionality of the unknown parameter vector, high convergence speed, realization simplicity and ability of the achieved accuracy estimation. The accuracy analysis is based on general parameter average and variance estimation during the learning procedure steady state (after GP statistics stabilization). The structure optimality criterion for the GP RBFN (General Parameter Radial Basis Function Network) is derived using the GP average and variance values. The derived criterion is used then for the development of the GP RBFN structure self-organization procedure. As a result, an Adaptive Fuzzy System (AFS) with capability to extract fuzzy If-Then rules from input and output sample data is proposed. Simulation examples are given.
View full abstract
-
1998 Volume 10 Issue 5 Pages
983-1008
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS
-
1998 Volume 10 Issue 5 Pages
1009-
Published: October 15, 1998
Released on J-STAGE: January 07, 2018
JOURNAL
FREE ACCESS