In this paper, we have proposed a new method for diffuse liver disease classification with sonogram, including the normal liver, hepatitis and cirrhosis, from a new point of view "scale. " The new system utilizes a multiscale analysis tool, called wavelet transforms, to analyze the ultrasonic liver images. A new set of features consisting of second order statistics derived from the wavelet transformed images is employed. From these features, we have found that the third scale is the representative scale for the classification of the considered liver diseases, and the horizontal wavelet transform can improve the representation of the corresponding features. Experimental results show that our method can achieve about 88% correct classification rate which is superior to other measures such as the co-occurrence matrices, the Fourier power spectrum, and the texture spectrum. This implies that our feature set can access the granularity from sonogram more effectively. It should be pointed out that our features are powerful for discriminating the normal livers from the cirrhosis because there is no misclassification samples between the normal liver and the cirrhosis sets. In addition, the experimental results also verify the usefulness of "scale" because our multiscale feature set can gain eighteen percent advantage over the direct use of the statistical features. This means that the wavelet transform at proper scales can effectively increase the distances among the statistical feature clusters of different liver diseases.

We propose a rough classification system using Hierarchical Learning Vector Quantization (HLVQ) for large scale classification problems which involve many categories. HLVQ of proposed system divides categories hierarchically in the feature space, makes a tree and multiplies the nodes down the hierarchy. The feature space is divided by a few codebook vectors in each layer. The adjacent feature spaces overlap at the borders. HLVQ classification is both speedy and accurate due to the hierarchical architecture and the overlapping technique. In a classification experiment using ETL9B, the largest database of handwritten characters in Japan, (it contains a total of 607,200 samples from 3036 categories) the speed and accuracy of classification by HLVQ was found to be higher than that by Self-Organizing feature Map (SOM) and Learning Vector Quantization methods. We demonstrate that the classification rate of the proposed system which uses multi-codebook vectors for each category under HLVQ can achieve higher speed and accuracy than that of systems which use average vectors.

The increasing diversity in Internet applications necessitates extended Internet architecture that can differentiate forwarding treatment of different types of flows. Diffserv can be a solution to the problem when it is augmented by several additional components. In this paper we describe various issues and possible directions in augmenting Diffserv. We present our analysis of Diffserv architecture, anticipated developments to augment Diffserv architecture, and potential applications of Diffserv.

In this paper, we present a class of combinatorial-logical classifiers called test feature classifiers. These are polynomial functions that can be used as pattern classifiers of binary-valued feature vectors. The method is based on so-called tests, sets of features, which are sufficient to distinguish patterns from different classes of training samples. Based on the concept of test we propose a new distance-based test feature classifiers. To test the performance of the classifiers, we apply them to a well-known phoneme database and to a textual region location problem where we propose a new effective textual region searching system that can locate textual regions in a complex background. Experimental results show that the proposed classifiers yield a high recognition rate than conventional ones, have a high ability of generalization, and suggest that they can be used in a variety of pattern recognition applications.

This paper presents a server-based approach to maintaining software integrity for all computer classroom workstations. The approach takes several advantages: (1) applicable to the FAT (file allocation table) and NTFS file systems, (2) renovating all workstations to workable state, (3) quickly adding or removing software systems to or from all workstations for teachers conducting new courses, and (4) automatically changing computer name and IP (Internet Protocol) address to an appointed one. The basic concept of the server-based maintenance approach is to install whole software systems, including operating system and applications, on a normal workstation, to make one image copy of the workstation's hard disk and store it onto network server, and to restore the image file from the server to the remaining workstations. In order to change computer name and IP automatically, this paper presents a searching heuristic for finding their locations in the image file. The heuristic is modified from Boyer-Moore (BM) algorithm and can improve the BM algorithm's performance over 9%.

Given N real weights w1, w2, . . . , wN stored in one-dimensional array, we consider the problem for finding an optimal interval I [1, N] under certain criteria. We shall review efficient algorithms developed for solving such problems under several optimality criteria. This problem can be naturally extended to two-dimensional case. Namely, given a NN two-dimensional array of N2 reals, the problem seeks to find a subregion of the array (e. g. , rectangular subarray R) that optimizes a certain objective function. We shall also review several algorithms for such problems. We shall also mention applications of these problems to region segmentation in image processing and to data mining.

A multilayer perceptron is usually considered a passive learner that only receives given training data. However, if a multilayer perceptron actively gathers training data that resolve its uncertainty about a problem being learnt, sufficiently accurate classification is attained with fewer training data. Recently, such active learning has been receiving an increasing interest. In this paper, we propose a novel active learning strategy. The strategy attempts to produce only useful training data for multilayer perceptrons to achieve accurate classification, and avoids generating redundant training data. Furthermore, the strategy attempts to avoid generating temporarily useful training data that will become redundant in the future. As a result, the strategy can allow multilayer perceptrons to achieve accurate classification with fewer training data. To demonstrate the performance of the strategy in comparison with other active learning strategies, we also propose an empirical active learning algorithm as an implementation of the strategy, which does not require expensive computations. Experimental results show that the proposed algorithm improves the classification accuracy of a multilayer perceptron with fewer training data than that for a conventional random selection algorithm that constructs a training data set without explicit strategies. Moreover, the algorithm outperforms typical active learning algorithms in the experiments. Those results show that the algorithm can construct an appropriate training data set at lower computational cost, because training data generation is usually costly. Accordingly, the algorithm proves the effectiveness of the strategy through the experiments. We also discuss some drawbacks of the algorithm.

We address the problem of computing various types of expressive tests for decision trees and regression trees. Using expressive tests is promising, because it may improve the prediction accuracy of trees, and it may also provide us some hints on scientific discovery. The drawback is that computing an optimal test could be costly. We present a unified framework to approach this problem, and we revisit the design of efficient algorithms for computing important special cases. We also prove that it is intractable to compute an optimal conjunction or disjunction.

Petri net is a graphical and mathematical modeling tool for discrete event systems. This paper treats analysis problems of time Petri nets. In this model, a minimal and a maximal firing delays are assigned to each transition. If a transition is 'enabled' it can fire after minimal delay has passed and must fire before maximal delay has elapsed. Since time Petri net can simulate register machines, it has equivalent modeling power to that of Turing machine. It means, however, that most of the analysis problems of time Petri nets with general net structures are undecidable. In this paper, net structures are restricted to a subclass called partially ordered condition (POC) nets and dissynchronous choice (DC) nets. Firing delays are also restricted to satisfy 'static fair condition' which assures chance to fire for all transitions enabled simultaneously. First, a sufficient condition of liveness of time POC net with the static fair condition is derived. Then it is shown that liveness of time DC net with static fair condition is equivalent to liveness of the underlying nontime net. This means that liveness problem of this class is decidable. Lastly, liveness problem of extended free choice (EFC) net is shown to be decidable.

In this paper, we investigate the bit error rate (BER) performance of Direct Sequence-Code Division Multiple Access (DS-CDMA) systems under impulsive radio noise environments, and propose a novel DS-CDMA receiver which is designed to be robust against impulsive noise. At first, employing the Middleton's Class-A impulsive noise model as a typical model of impulsive radio noise, we discuss the statistical characteristics of impulsive radio noise and demonstrate that the quadrature components of impulsive noise are statistically dependent. Next, based on the computer simulation, we evaluate the BER performance of a conventional DS-CDMA system under a Class-A impulsive noise environment, and illustrate that the performance of the conventional DS-CDMA system is drastically degraded by the effects of the impulsive noise. To deal with this problem, motivated by the statistical dependence between the quadrature components of impulsive radio noise, we propose a new DS-CDMA receiver which can eliminate the effects of the channel impulsive noise. The numerical result shows that the performance of the DS-CDMA system under the impulsive noise environment is significantly improved by using this proposed receiver. Finally, to confirm the effectiveness of this proposed receiver against actual impulsive radio noise, we evaluate the BER performance of the DS-CDMA system employing the proposed receiver under a microwave oven (MWO) noise environment and discuss the robustness of the proposed receiver against MWO noise.

In order to investigate the dynamic behavior of quantized interconnection neural networks on neuro-chips, we have designed and fabricated hardware neural networks according to design rule of a 1.2 µm CMOS technology. To this end, we have developed programmable synaptic weights for the interconnection with three values of 1 and 0. We have tested the chip and verified the dynamic behavior of the networks in a circuit level. As a result of our study, we can provide the most straightforward application of networks for a dynamic pattern classifier. The proposed network is advantageous in that it does not need extra exemplar to classify shifted or reversed patterns.

The self-healing capability against network failure is one of indispensable features for the B-ISDN infrastructure. One problem in realizing such self-healing backbone network is the inefficient utilization of the large spare capacity designed for the failure-restoration purpose since it will be used only in the failure time that does not occur frequently. "Pool-capacity" is the concept that allows some VPs (virtual paths) to efficiently utilize this spare capacity part. Although the total capacity can be saved by using the "Pool Capacity," it is paid by less reliability of VPs caused by the emerging influence of indirect-failure. Thus, this influence of indirect-failure has to be considered in the capacity designing process so that network-designers can trade off the saving of capacity with the reliability level of VPs in their self-healing networks. In this paper, Damage Rate:DR which is the index to indicate the level of the influence caused by indirect-failure is defined and the pool-capacity design scheme with DR consideration is proposed. By the proposed scheme, the self-healing network with different cost (pool-capacity) can be designed according to the reliability level of VPs.

Feedback of class memberships is incorporated into multimodal pattern classifiers and their unsupervised learning algorithm is presented. Classification decision at low levels is revised by the feedback information which also enables the reconstruction of patterns at low levels. The effects of the feedback are examined for the McGurk effect by using a simple model.

In this paper, a classified vector quantization (CVQ) method using a novel direction based classifier is proposed. The new classifier uses a distortion measure related to the angle between vectors to determine the similarity of vectors. The distortion measure is simple and adequate to classify various edge types other than single and straight line types, which limit the size of image block to a rather small size. Simulation results show that the proposed technique can achieve better perceptual quality and edge integrity at a larger block size, as compared to other CVQs. It is shown when the vector dimension is changed from 16(4 4) to 64(8 8), the average bit rate can be reduced from 0. 684 bpp to 0.191, whereas the PSNR degradation is only about 1.2 dB.

We apply two acceleration techniques for the backpropagation algorithm to an iterative gradient descent algorithm called the network inversion algorithm. Experimental results show that these techniques are also quite effective to decrease the number of iterations required for the detection of input vectors on the classification boundary of a multilayer perceptron.

This paper describes a way of implementing a conceptual model for temporal data on a commercial object database system. The implemented version is provided as a class library. The library enables applications to handle temporal data. Any application can employ the library because it does not depend on specific applications. Furthermore, we propose an enhanced version of Time Index. The index efficiently processes event queries in particular. These queries search time intervals in which given events are all valid. We also investigate the effectiveness of the enhanced Time Index.

A major topic in the field of network and telecommunications is doing business on the World Wide Web (WWW), which is called Electronic Commerce (EC). Another major topic is blending Artificial Intelligence (AL) techniques with the WWW. In the Ex-W-Pert Project we have proposed an agent model for EC components that blends the traditional expert systems' reasoning engine with a multi-layer knowledge base, communication and documentation engines. In this project, EC is viewed as a society of software agents, such as customer, search, catalog, manufacturer, dealer, delivery and banker agents, interacting and negotiating with each other. Each agent has a knowledge-base and a reasoning engine, a communication engine and a documentation engine. The knowledge-base is organized in three layers: skill layer, rule layer and knowledge layer (S-R-K layers). In this project, for each EC agent, we identify the class of problems to be solved and build the knowledge base gradually for each layer. We believe that using this multi-layer knowledge base system will speed up the reasoning and ultimately reduce the operation costs.

In this paper, we represent a class structure using directed graph in which each node corresponds to each member of the class. To quantify the dependence relationship among members, we define weighted closure. Using this quantified relationship and effort equation proposed by M. H. Halstead, we propose a metric for class structural complexity.

An automatic defect classification system (ADC) for use in visual inspection of semiconductor wafers is introduced. The methods of extracting the defect features based on the human experts' knowledge, with their correlations with the defect classes are elucidated. As for the classifier, Hyperellipsoid Clustering Network (HCN) which is a layered network model employing second order discrimination borders in the feature space, is introduced. In the experiments using a collection of defect images, the HCNs are compared with the conventional multilayer perceptron networks. There, it is shown that the HCN's adaptive hyperellipsoidal discrimination borders are more suited for the problem. Also, the cluster encapsulation by the hyperellipsoidal border enables to determine rejection classes, which is also desirable when the system will be in actual use. The HCN with rejection achieves, an overall classification rate of 75% with an error rate of 18%, which can be considered equivalent to those of the human experts.

Image classification is an important task in document image analysis and understanding, page segmentation-based document image compression, and image retrieval. In this paper, we present a new approach for distinguishing textual images from pictorial images using the Kolmogorov Complexity (KC) measure with randomly extracted blocks. In this approach, a number of blocks are extracted randomly from a binarized image and each block image is converted into a one-dimensional binary sequence using either horizontal or vertical scanning. The complexities of these blocks are then computed and the mean value and standard deviation of the block complexities are used to classify the image into textual or pictorial image based on two simple fuzzy rules. Experimental results on different textual and pictorial images show that the KC measure with randomly extracted blocks can efficiently classified 29 out 30 images. The performance of our approach, where an explicit training process is not needed, is comparable favorably to that of a neural network-based approach.