C1 class smooth interpolation by a fuzzy reasoning for a small data set is proposed. The drafting technique of a human expert is implemented by using a set of fuzzy rules. The effectiveness of the present method is verified by computer simulations and by applications to the practical interpolation problem in a power system.

For reducing requirement of large memory and minimizing computation complexity in a large-vocabulary continuous speech recognition system, speech segmentation plays an important role in speech recognition systems. In this paper, we formulate the speech segmentation as a two-phase problem. Phase 1 (frame labeling) involves labeling frames of speech data. Frames are classified into three types: (1) silence, (2) consonant and (3) vowel according to two segmentation features. In phase 2 (syllabic unit segmentation) we apply the concept of transition states to segment continuous speech data into syllabic units based on the labeled frames. The novel class of hyperrectangular composite neural networks (HRCNNs) is used to cluster frames. The HRCNNs integrate the rule-based approach and neural network paradigms, therefore, this special hybrid system may neutralize the disadvantages of each alternative. The parameters of the trained HRCNNs are utilized to extract both crisp and fuzzy classification rules. In our experiments, a database containing continuous reading-rate Mandarin speech recorded from newscast was utilized to illustrate the performance of the proposed speaker independent speech segmentation system. The effectiveness of the proposed segmentation system is confirmed by the experimental results.

This paper presents a method for mechanically transforming a parallel algorithm on an original network so that the algorithm can work on a target network. It is assumed that the networks are of cube-type such as the shuffle-exchange network, omega network, and hypercube. Were those networks isomorphic to each other, the algorithm transformation is an easy task. The proposed transformation method is based on a novel graphembedding scheme <φ: δ, κ, π, ψ>. In addition to the dilating operation δ of the usual embedding scheme <φ: δ>, the novel scheme uses three primitive graph-transformation operations; κ (= δ-1) for contracting a path into a node, π for pipelining a graph, and ψ (= π-1) for folding a pipelined graph. By applying the primitive operations, the cube-type networks can be transformed so as to be isomorphic to each other. Relationships between the networks are represented by the composition of applied operations. With the isomorphic mapping φ, an algorithm in a node of the original network can be simulated in the corresponding node(s) of the target network. Thus the algorithm transformation is reduced to routine work.

For a fuzzy control of manipulated variable so as to match a required output of a plant, tuning of fuzzy rules are necessary. For its purpose, various methods to tune their rules automatically have been proposed. In these method, some of them necessitate much time for its tuning, and the others are lacking in the generalization capability. In the fuzzy control by the steepest descent method, a use of piecewise linear membership functions (MSFs) has been proposed. In this algorithm, MSFs of the premise for each fuzzy rule are tuned having no relation to the other rules. Besides, only the MSFs corresponding to the given input and output data for the learning can be tuned efficiently. Comparing with the conventional triangular form and the Gaussian distribution of MSFs, an expansion of the expressiveness is indicated. As a result, for constructing the inference rules, the training cycles can be reduced in number and the generalization capability to express the behavior of a plant is expansible. An effectiveness of this algorithm is illustrated with an example of a parallel parking of an autonomous mobile robot.

A new cost model CPO (Cost of Process Ownership) has been proposed. We have already the well known cost model CEO (Cost of Equipment Ownership) [1] which is a cost index assigned independently to individual equipment. However, CPO is basically a cost index assigned to process step in processing flow chart of actual product in a Fab line. Therefore, it is essentially more effective to evaluate the Fab performance such as cost analysis of process steps, estimation of the whole wafer processing cost for specific product, identification of the bottle neck process step or equipment in a Fab line. Further, in designing a high cost-performance factory or in modifying existing factory, it affords important guide such as optimal scales for both factory and equipments with their investment efficiency.

This paper proposes a Japanese continuous speech recognition mechanism in which a full-sentence-level context-free-grammar (CFG) and one kind of semantic constraint called dependency relationships between two bunsetsu (a kind of phrase) in Japanese" are used during speech recognition in an integrated way. Each dependency relationship is a modification relationship between two bunsetsu; these relationships include the case-frame relationship of a noun bunsetsu to a predicate bunsetsu, or adnominal modification relationships such as a noun bunsetsu to a noun bunsetsu. To suppress the processing overhead caused by using relationships of this type during speech recognition, no rigorous semantic analysis is performed. Instead, a simple matching with examples" approach is adopted. An experiment was carried out and results were compared with a case employing only CFG constraints. They show that the speech recognition accuracy is improved and that the overhead is small enough.

This paper discusses an object-oriented approach to temporal multimedia data modeling in OMEGA; a multimedia database management under development at the University of Library and Information Science. An object-orientated approach is necessary to integrate various types of heterogeneous multimedia data, but it has become clear that current object-oriented data models are not sufficient to represent multimedia data, particularly when they are temporal. For instance, the current object-oriented data models cannot describe objects whose attribute values change time-dependently. Also, they cannot represent temporal relationships among temporal multimedia objects. We characterize temporal objects as instances of a subclass of class TimeInterval with the temporal attributes and the temporal relationships. This temporal multimedia data model is designed upward compatible with the ODMG-93 standard object model. To organize a temporal multimedia database, a five temporal axes model for representing temporal multimedia objects is also introduced. The five temporal axes--an absolute, an internal, a quasi-, a physical, and a presentation time axis--are necessary to describe time-dependent properties of multimedia objects in modeling, implementing and use. A concrete example of this organization method is also illustrated.

Recognition of hand drawn shapes is beneficial in drawing packages and automated sketch entry in hand-held computers. Although it is possible to store and retrieve drawings through the use of electronic ink, further manipulation of these drawings require recognition to be performed. In this paper, we propose a new approach to invariant geometric shape recognition which utilizes a fuzzy function to reduce noise and a neural network for classification. Instead of recognizing segments of a drawing and then performing syntactical analysis to match with a predefined shape, which is weak in terms of generalization and dealing with noise, we examine the shape as a whole. The main concept of the recognition method is derived from the fact that internal angles are very important in the perception of the shape. Our application's aim is to recognize and correctively redraw hand drawn ellipses, circles, rectangles, squares and triangles. The neural network learns the relationships between the internal angles of a shape and its classification, therefore only a few training samples which represent the class of the shape is sufficient. The results are very successful, such that the neural network correctly classified shapes which were not included in the training set.

Managed Objects (MOs) can be specified by the combination of a static information model and a dynamic process model. First, this paper presents a mapping of attributes from a process model diagram to an information model diagram. Then, it introduces a concept of topology into these two models and proposes a hypothesis about the relationship of topology in these two models. To explicitly explain the hypothesis, it can be stated that all attributes of incoming or outgoing data related to a process in a process model are mapped to an information model where these attributes are interconnected by an explicit relationship which corresponds to a specific meaning, such as physical containment or logical connection. From an intuitive perspective, it can be said that if two attributes have a close topological relationship in a process model, the mapped attributes also have a close topological relationship in an information model. This hypothesis provides clues for determining whether there is an error in an attribute either in the process model or in the information model. By examining the way attributes of incoming or outgoing data related to a process are mapped to an information model, we can detect whether there is an error with respect to the process. The error correction is performed with the assistance of probability analysis. The method of error detection and correction can be implemented in a computer aided tool. Then, error detection on the attribute level becomes automatic, and error correction on the attribute level becomes interactive through the computer aided tool. Finally, the validity of the hypothesis is confirmed by analyzing ITU-T Recommendation M.3100. The specification of the fabric object class defined in M.3 100 is transformed into these two models and the hypothesis is validated for the analysis of the mapping between these two models.

The Hyper Hospital" is a novel medical care system which will be constructed on an electronic information network. The human interface of the Hyper Hospital based on the modern virtual reality technology is expected to enhance patients' ability to heal by providing computer-supported on-line visual consultations. In order to investigate the effects and features of on-line visual consultations in the Hyper Hospital, we conducted an experiment to clarify the influence of electronic interviews on the talking behavior of interviewees in the context of simulated doctor-patient interactions. Four types of distant-confrontation interviews were made with voluntary subjects and their verbal and non-verbal responses were analyzed from the behavioral point of view. The types of interviews included three types of electronic media-mediated interviews and one of a live face to face interview. There was a tendency in the media-mediated interviews that both the latency and the duration of interviewees' utterances in answering questions increased when they were compared with those of live face to face interviews. These results suggest that the interviewee became more verbose or talkative in the media-mediated interviews than in the live interviews. However, the interviewee's psychological tension was generally augmented in the media-mediated interviews, which was suggested by the delay of the initiation of conversations as compared to the conventional face-to-face interviews. We also discuss the applicability of media-mediated interviews by an electronic doctor which we are studying as a functional unit of our Hyper Hospital, a network based virtual reality space for medical care.

A fuzzy inference processor which performs fuzzy inference with 12-bit resolution input at 200 kFLIPS (Fuzzy Logical Inference Per Second) has been developed. To keep the cost performance, not parallel processing hardware but processor type hardware is employed. Dedicated membership function generators, rule instructions and modified add/divide algorithm are adopted to attain the performance. The membership function generators calculate a membership function value in less than a half clock cycle. Rule instructions calculate the grade of a rule by one instruction. Antecedent processing and consequent processing are pipelined by the modified add/divide algorithm. As a result, total inference time is significantly reduced. For example, in the case of typical inference (about 20 rules with 2 to 4 inputs and 1 output), the total inference needs approximately 100 clock cycles. Furthermore by adding a mechanism to calculate the variance and maximum grade of the final membership function, it is enabled to evaluate the inference reliability. The chip, fabricated by 1 µm CMOS technology, contains 86k transistors in a 7.56.7 mm die size. The chip operates at more than 20 MHz clock frequency at 5 V.

Given an integer N, it is easy to determine whether or not N is prime, because a set of primes is in LPP. Then given a composite number N, is it easy to determine whether or not N is of a specified form? In this paper, we consider a subset of odd composite numbers +1MOD4 (resp. +3MOD4), which is a subset of odd composite numbers consisting of prime factors congruent to 1 (resp. 3) modulo 4, and show that (1) there exists a four move (blackbox simulation) perfect ZKIP for the complement of +1MOD4 without any unproven assumption; (2) there exists a five move (blackbox simulation) perfect ZKIP for +1MOD4 without any unproven assumption; (3) there exists a four move (blackbox simulation) perfect ZKIP for +3MOD4 without any unproven assumption; and (4) there exists a five move (blackbox simulation) statistical ZKIP for the complement of +3MOD4 without any unproven assumption. To the best of our knowledge, these are the first results for a language L that seems to be not random self-reducible but has a constant move blackbox simulation perfect or statistical ZKIP for L and without any unproven assumption.

In this paper, we propose a multi-purpose proof system which enables a user remembering only one piece of secret data to perform various proof protocols. These proofs include identity proof, membership proof without disclosing identity, and combined identity and membership proof. When a user participates in a group, he will obtain a secret witness from the group administrator. Many secret witnesses can be combined into one piece of secret data. But the size of the secret data is independent of the number of the groups in which the user participates. Our system satisfies other desirable properties which were not attained by the previously proposed systems.