A new fuzzy inference model, "SIRMs(Single Input Rule Modules) Connected Fuzzy Inference Model", for plural input fuzzy control is proposed. In the model, the degree of importance is defined first and single input fuzzy rule module is constructed for each input item. Then, the final output is obtained by summarizing the products of the importance degree and the fuzzy inference result of each module. When the number of input items increases, the total number of fuzzy rules rises only algebraically for the proposed model while ascends exponentially for the conventional model. Moreover, the importance degrees can be strengthened or weakened according to experts' intuitive experiences to achieve each control purpose. The proposed model is applied to typical first-order lag systems and second-order lag systems to confirm the improvement in control performance compared with the conventional models. Tuning algorithm is also given based on the simplified inference method. Finally, the proposed model is applied to identification of non-linear functions of 4 inputs. The results show that the proposed model has the ability to identify non-linear systems, too.

View full abstract

Rapid progress of information services has been increasing the amount of text information which users receive on Internet, teletext broadcasting and so on. A system extracting important part automatically is useful for users to get their interest information easily in such a case. Some methods in the field of natural language processing can extract important keywords or phrases from text information by using word dictionaries. However, these methods can hardly deal with unexpected words such as new proper nouns.On the other hand, the methods based on word occurrence are useful for dealing with unexpected words and they have high applicability for many kinds of text information because they use no dictionary. However, the results are insufficient as a summary in some cases.In this paper, we describe our newly developed system KEIFIS (Keyword Extracting and Information FIltering System). KEIFIS extracts important keywords which represent major topics from a large amount of Japanese text information and collects related data with no dictionary.KEIFIS has the following features : (1) It automatically extracts important keywords and combines some of them to give the users major topics.(2) It retrieves and collects information according to the topic specified by the users, and informs them of its arrival in real time.KEIFIS employs fuzzy information processing in computing the similarity between words. The similarity is utilized in counting co-occurrence of the related words to extract the major topics in KEIFIS. The similarity is also used to define relationship between the specified topic and newly provided information.We applied KEIFIS to news programs on the teletext broadcasting in Japanese, and made sure that KEIFIS was capable of extracting the important keywords which represent current major topics appropriately.

View full abstract

This paper proposes a new systematic approach of analyzing and recognizing difficult Braille characters. The approach is based on integrating image processing and fuzzy reasoning, where fuzzy reasoning played key role in formulating the classification algorithm, and image processing in identifying effective features to be used in the fuzzy rule evaluation. We introduced five features(observable) on the amounts of bright-or-dark-pixels (T), highlighted-pixels (H), shaded-pixels (S), pixels included in area which boundary is defined by sharp change in the brightness (C), and pixels included in boundary area (E), and to each of which membership functions are introduced to represent deformed Braille-dot sets of "worn-down" "stained", "cracked", "holed", and "normal", that are mapped to membership functions defined to the feature variables space respctively. The fuzzy clustering analysis was used to find the five feature parameters, and the membership functions were trained to 5600 teacher data points. The new recognition approach showed very high accuracy of 99.99%. Since our approach recognizes also the types of deformed Braille character, it fines quality level of legacy books easily for recommending whether refreshing should be made or not to refresh them.

View full abstract

Experimental and computational fluid dynamics(CFD) approaches are now orthodox means in flow prediction. However, in some engineering applications where these approaches are used as a design tool, their expensive and time-consuming nature may hamper the process of reaching specialists' final goal. In this study, we investigated the potential of applying structured artificial neural networks to fluid dynamical problems. A typical hydraulic flow phenomenon, the Karman vortex street was examined here. For realizing the reasoning procedure in this investigation, the sensitivity study of the horizontal velocity profiles on several cross sections over the flow field was conducted. Based on the sensitivity study, three structured neural networks were employed to carry out the flow pattern estimation. They were modeled as the action side of a qualitative rule to work in the reasoning procedure. Compared with the computational fluid dynamical solutions, the estimation accuracy is very encouraging. Furthermore, the proposed reasoning procedure can give a prompt answer compared with those time-consuming conventional approaches.

View full abstract

Cognitive maps and time space maps make use of multidimensional scaling (MDS) techniques to analyze data relating to spatial and environmental preferences and perception. Perceptual configuration of the points is represented by a cognitive map with surface feature interpolation. In this paper we propose a procedure of MDS using Neurofuzzy adaptive modelling with B-splines for surface feature interpolation. The procedure is compared with those using Hayashi's quantifying method of paired comparisons and Torgerson's metrical multidimensional scaling procedure. In numerical examples, the resultant maps are compared and an application to sociometry analysis is presented. It is shown that by introducing a simple term to the objective function the topological distortion can be drastically improved.

View full abstract

In this paper, we focus on large-scale multiobjective fuzzy linear programming problems with the block angular structure and examine the efficiency of the DAntzig-Wolfe decomposition method in the interactive fuzzy satisficing method recently proposed by Sakawa et al. After overviewing the Dantzig-Wolfe decomposition method and the interactive fuzzy satisficing method, three-objective linear programming problems with 15 coupling constraints are considered in order to demonstrate the efficiency of the Dantzig-Wolfe decomposition method over the revised simplex method. Through a lot of computational experiments on workstation for numerical examples with both 50 and 200 variables, the advantages of the Dantzig-Wolfe decomposition method are discussed with respect to processing time and required memory storage.

View full abstract

A production function is to illustrate the relation between the products and producing activities in an industry. Although it is required that coefficient values of the production function should be stable, the stable values of a coefficient is rarely obtained out of statistics of changing economic phenomena.The change of the coefficient values included in the production function can be interpreted to show the possibilities of economy in some sense. In other words, it is natural that the capital input and the labor population should change under a changeable economic state. This causes the change of coefficient values. In this paper, the stress should be placed on that possibilistic regression analysis can be employed to analyze industries considering the possibility of economy. And as an application of the production function, the features of Japanese and American industries are successfully analyzed and the characteristics of Japanese industries are explained by analyzing their cross section data.

View full abstract

The evaluation of efficiency for DMU_s(Decision Making Units) in some company is very important. In this view point, the study of evaluation of efficiency is very active, on the basis of the formulation for the production function. Recently, DEA (Data Envelopment Analysis) is focused on the evaluation of efficiency. However, we often obtain DEA efficiency for DMU_o from the efficiency of one output. This means DEA analyzes one output by some inputs and ignores another outputs. Furthermore, DEA efficiency is bigger than the max efficiency among one object solutions for an individual output. From above-mentioned two problem in DEA, as we need to obtain the total efficiency analyzed from all outputs by all inputs, it is natural that the total efficiency exists between min efficiency and max efficiency among one object solutions for an individual output. In this view point, we try to introduce the concept of a Fuzzy Goal into DEA. But, the satisfactional efficiency is always the max efficiency, because the max efficiency is in the possibility solution area with satisfactional rate which is one. Therefore, as DEA efficiency is the linear weighted sum of individual efficiency for every output, we try to change DEA model to the multi-objective problem, and propose the satisfactional mehtod for DUM_o along the maximizing decision defined by Bellman and Zadeh, and propose the improvement method for the satisfactinal solution along the dialogue to DM by trad/off rate for two efficiencies which is developed by Sakawa.

View full abstract