Search

In this paper we propose a novel personalized paper search system using the relevance among user's queried keywords and user's behaviors on a searched paper list. The proposed system builds user's individual relevance network from analyzing the appearance frequencies of keywords in the searched papers. The relevance network is personalized by providing weights to the appearance frequencies of keywords according to users' behaviors on the searched list, such as "downloading," "opening," and "no-action." In the experimental section, we demonstrate our method using 100 faculties' search information in the University of Suwon. [ABSTRACT FROM AUTHOR]

This paper is concerned with a multi-resolution tool for screening paper formation variations in various frequency regions on production line. A paper web is illuminated by two red diode lasers and the reflected light recorded as two time series of high resolution measurements constitute the input signal to the papermaking process monitoring system. The time series are divided into blocks and each block is analyzed separately. The task is treated as kernel based novelty detection applied to a multi-resolution time series representation obtained from the band-pass filtering of the Fourier power spectrum of the series. The frequency content of each frequency region is characterized by a feature vector, which is transformed using the canonical correlation analysis and then categorized into the inlier or outlier class by the novelty detector. The ratio of outlying data points, significantly exceeding the predetermined value, indicates abnormalities in the paper formation. The tools developed are used for online paper formation monitoring in a paper mill. [ABSTRACT FROM AUTHOR]

This paper is concerned with a multi-resolution tool for analysis of a time series aiming to detect abnormalities in various frequency regions. The task is treated as a kernel based novelty detection applied to a multi-level time series representation obtained from the discrete wavelet transform. Having a priori knowledge that the abnormalities manifest themselves in several frequency regions, a committee of detectors utilizing data dependent aggregation weights is build by combining outputs of detectors operating in those regions. [ABSTRACT FROM AUTHOR]

The techniques for organizing and retrieving the artifacts from digital libraries (DLs) semantically are discussed, which include letting the taxonomies and semantic relations work in tandem to index the artifacts in DLs; integrating the techniques used in natural language processing and taxonomies to help users to start their retrieval processes; and ranking scientific papers on similarity in terms of contents or ranking the relevant papers on multi-factors. These techniques are verified through the design and implementation of a prototype of DLs for scientific paper management. [ABSTRACT FROM AUTHOR]

The aim of this paper is to give theoretical and experimental tools for measuring the driving force in evolving complex networks. First a discrete-time stochastic model framework is introduced to state the question of how the dynamics of these networks depend on the properties of the parts of the system. Then a method is presented to determine this dependence in the possession of the required data about the system. This measurement method is applied to the citation network of high energy physics papers to extract the in-degree and age dependence of the dynamics. It is shown that the method yields close to "optimal" results. [ABSTRACT FROM AUTHOR]

Nowadays amount of information on Internet is dramatically increasing. The ability of facilitating users to achieve useful information is more and more important for search field. CDSE, a model for the domain-based intelligent search engine is proposed in this paper. The model can help users to retrieve what they need by combining text classification with keywords extraction. Several algorithms that use key technologies are proposed, such as statistics, data mining and agents. Then a new criterion named ranking error is contributed to solve the problem of evaluation ranking inefficiency in traditional performance evaluation methodologies. The experimental results indicate that the proposed model can effectively improve retrieval precision and solve the problem of relevant document ranking behind in current search engine. [ABSTRACT FROM AUTHOR]

Reading and literature exploration are important tasks of scientific research. However, conventional retrieval systems provide limited support for these tasks by concentrating on identifying relevant materials. New generation systems should provide additional support functionality by focusing on analyzing and organizing the retrieved materials. A framework of literature exploration support systems is proposed. Techniques of granular computing are used to construct granular knowledge structures from the contents, structures, and usages of scientific documents. The granular knowledge structures provide a high level understanding of scientific literature and hints regarding what has been done and what needs to be done. As a demonstration, we examine granular knowledge structures obtained from an analysis of papers from two rough sets related conferences. [ABSTRACT FROM AUTHOR]

In this paper we propose a two-pass classification method and apply it to face recognitions. The method is obtained by integrating together two approaches, the hyper-ellipsoid neural networks (HENN's) and the SVM's with error correcting codes. This method realizes a classification operation in two passes: the first one is to get an intermediate classification result for an input sample by using the HENN's, and the second pass is followed by using the SVM's to re-classify the sample based on both the input data and the intermediate result. Simulations conducted in the paper for applications to face recognition showed that the two-pass method can maintain the advantages of both the HENN's and the SVM's while remedying their disadvantages. Compared with the HENN's and the SVM's, a significant improvement of recognition performance over them has been achieved by the new method. [ABSTRACT FROM AUTHOR]

This paper proposed an automatic recognition system of shipping container identifiers using fuzzy-based noise removal method and ART2-based self-organizing supervised learning algorithm. Generally, identifiers of a shipping container have a feature that the color of characters is black or white. Considering such a feature, in a container image, all areas excepting areas with black or white colors are regarded as noises, and areas of identifiers and noises are discriminated by using a fuzzy-based noise detection method. Noise areas are replaced with a mean pixel value of the whole image and areas of identifiers are extracted by applying the edge detection by Sobel masking operation and the vertical and horizontal block extraction in turn to the noise-removed image. Extracted areas are binarized by using the iteration binarization algorithm, and individual identifiers are extracted by applying 8-directional contour tacking method. This paper proposed an ART2-based self-organizing supervised learning algorithm for the identifier recognition, which creates nodes of the hidden layer by applying ART2 between the input and the hidden layers and improves the performance of learning by applying generalized delta learning and Delta-bar-Delta algorithm between the hidden and the output layers. Experiments using many images of shipping containers showed that the proposed identifier extraction method and the ART2-based self-organizing supervised learning algorithm are more improved compared with the methods previously proposed. [ABSTRACT FROM AUTHOR]

Accurate cancer classification is very important to cancer diagnosis and treatment. As molecular information is increasing for the cancer classification, a lot of techniques have been proposed and utilized to classify and predict the cancers from gene expression profiles. In this paper, we propose a method based on speciated evolution for the cancer classification. The optimal combination among several feature-classifier pairs from the various features and classifiers is evolutionarily searched using the deterministic crowding genetic algorithm. Experimental results demonstrate that the proposed method is more effective than the standard genetic algorithm and the fitness sharing genetic algorithm as well as the best single classifier to search the optimal ensembles for the cancer classification. [ABSTRACT FROM AUTHOR]

Increasing the smooth degree of data series is key factor of grey model's precision. In this paper, the more general method is put forward on the basis of summarizing several kinds of ways to improve smooth degree of data series, and a new transformation is represented. The practical application shows the effectiveness and superiority of this method. [ABSTRACT FROM AUTHOR]

It presents an approach to the automatic generation of pencil sketching with the effects of paper texture. First, proposes a texture mapping of strokes in the aspect of skeleton, the filter algorithm and the standard deviation algorithm for rendering image, the near distances recover algorithm for real-time browsing, and finally implement the Pen-and-Ink Style of pencil sketching Rendering System. In the static rendering, it needn't adjust the threshold for the convenient on rendering and the outlines are more precise and exact. Besides, attaching the graftal and sketching shadow makes the composition of an image more attractive. Through a series of clever image processes, the system finally presents excellent colored pencil style drawings. Because the proposed algorithm is not complicated, the rendering time is quite short compared to other past related studies. Compared with other research works and Photoshop on a set of benchmarks, the system demonstrates its strength in the aspects of full automation, stability of sketching quality and higher visual satisfaction, all achieved in a considerably shorter time. [ABSTRACT FROM AUTHOR]

The storage capacity of the conventional neural network is 0.14 times of the number of neurons (P=0.14N). Due to the huge difficulty in recognizing large number of images or patterns,researchers are looking for new methods at all times. Quantum Neural Network (QNN), which is a young and outlying science built upon the combination of classical neural network and quantum computing,is a candidate to solve this problem.This paper presents Quantum Probability Distribution Network (QPDN) whose elements of the storage matrix are distributed in a probabilistic way on the base of quantum linear superposition and applies QPDN on image recognition. Contrasting to the conventional neural network, the storage capacity of the QPDN is increased by a factor of 2N,where N is the number of neurons. Besides,the case analysis and simulation tests have been carried out for the recognition of images in this paper, and the result indicates that QPDN can recognize the images or patterns effectively and its working process accords with quantum evolvement process. [ABSTRACT FROM AUTHOR]

This paper proposes a novel algorithm for nonnegative independent component analysis, which is based on minimizing the mutual information of the separated signals, and is truly insensitive to the particular underlying distribution of the source data. The unmixing system culminates to a novel neural network model. Compared with other algorithms for nonnegative ICA, the method proposed in this paper can work efficiently even in the case that the source signals are not well grounded, and that pre-whiting process is not needed. Finally, the experiments were performed on both simulating signals and mixtures of image data, the results indicate that the algorithm is efficient and effective. [ABSTRACT FROM AUTHOR]

In this paper we propose a formal approach to the generative design of artefacts. The founding idea, bridging the gap between the domain of architectural artefacts and the field of ontologies, is to represent the notion of species as it exists in the context of generative design by the concept of class existing in the field of formal ontologies. In this paper we propose an artefact design methodology and a system based on it that assists a designer in the building process of artefact 3D models. The implemented system is structured in three conceptual levels: the Design Strategy, the Computational Model, and the 3D graphical rendering system. The Computational model has been defined and developed using the Frame Logic formalism. The proposed methodology is independent of specific modelling environments. In this paper we also present methodological directions to couple artefact representations with 3D rendering systems. [ABSTRACT FROM AUTHOR]

This paper presents an approach for solving the driving load forecasting problem based on Cascade Neural Networks with node-decoupled extended Kalman Filtering (CNN-NDEKF). Because of the inherent advantages, hybrid electric vehicles (HEV) are being given more and more attention. The power control strategy of HEVs is the key technology which determines the HEV's efficiency and pollutive emission level. Since the extent of improvement involved with HEV power control strategies greatly depends on the future driving load forecasting, in this paper, we attempt to achieve driving load forecasting using CNN-NDEKF. Instead of forecasting the entire load sequence, we define 5 load levels by a fuzzy logic method and then we forecast the load level. Simulation study is given to illustrate the feasibility of the driving load forecasting approach. [ABSTRACT FROM AUTHOR]

Aiming at improving the convergence performance of conventional BP neural network, this paper presents an improved PSO algorithm instead of gradient descent method to optimize the weights and thresholds of BP network. The strategy of the algorithm is that in each iteration loop, on every dimension d of particle swarm containing n particles, choose the particle whose velocity decreases most quickly to mutate its velocity according to some probability. Simulation results show that the new algorithm is very effective. It is successful to apply the algorithm to gas turbine fault diagnosis. [ABSTRACT FROM AUTHOR]

In this paper, we study the Cohen-Grossberg neural networks with mixed time-varying delays. By applying the Lyapunov functional method and combining with the inequality 3abc ≤ a3 + b3 + c3 (a,b,c > 0) technique, a series of new and useful criteria on the existence of equilibrium point and its global asymptotical stability are established. The results obtained in this paper extend and generalize the corresponding results existing in previous literature. [ABSTRACT FROM AUTHOR]

This paper studies a general class of neural networks with time-varying delays and the neuron activations belong to the set of discontinuous monotone increasing functions. The discontinuities in the activations are an ideal model of the situation where the gain of the neuron amplifiers is very high. Because the delay in combination with high-gain nonlinearities is a particularly harmful source of potential instability, in the paper, conditions which ensure the global convergence of the neural network are derived. [ABSTRACT FROM AUTHOR]

Electromagnetic (EM)-based neural network (NN) approaches have recently gained recognition as unconventional and useful methods for radio frequency (RF) components modeling. In this paper, several improvement techniques including a new data preprocessing technique and an improved training algorithm are presented. Comprehensive cases are compared in this paper. The experimental results indicate that with these techniques, the modified model has better performance. [ABSTRACT FROM AUTHOR]

This paper addresses a capacitated production planning (CPP) problem with outsourcing. Specially, the outsourcing problem considered in this paper has the following features: (1) all demands are met by production, inventory, outsourcing, and sale loss without postponement or backlog; (2) The goal of the solution to this problem is to minimize the total cost covering the production, setup, inventory holding, sale loss, and outsourcing activities; (3) different from previous studies which usually involved some special assumptions on the cost function for these activities, the cost function structure we consider in this paper are arbitrary and time-varying; and (4) we consider a general problem situation where the production, inventory, and outsourcing have a capacity limitation. It is likely that there is no solution in the model due to the consideration of such general cost functions and many constraints, hence we employ a Bellman/Ford algorithm for checking feasible solutions of the problem and resulting in a feasible solution region. After that, a proposed algorithm is applied for searching a global optimal solution with in the feasible solution region. [ABSTRACT FROM AUTHOR]

The use of an ontology in order to provide a mechanism to enable machine reasoning has continuously increased during the last few years. This paper proposed an automated method for document classification using an ontology, which expresses terminology information and vocabulary contained in Web documents by way of a hierarchical structure. Ontology-based document classification involves determining document features that represent the Web documents most accurately, and classifying them into the most appropriate categories after analyzing their contents by using at least two pre-defined categories per given document features. In this paper, Web documents are classified in real time not with experimental data or a learning process, but by similarity calculations between the terminology information extracted from Web documents and ontology categories. This results in a more accurate document classification since the meanings and relationships unique to each document are determined. [ABSTRACT FROM AUTHOR]

This paper suggests an automated method for document classification using an ontology, which expresses terminology information and vocabulary contained in Web documents by way of a hierarchical structure. Ontologybased document classification involves determining document features that represent the Web documents most accurately, and classifying them into the most appropriate categories after analyzing their contents by using at least two pre-defined categories per given document features. In this paper, Web documents are classified in real time not with experimental data or a learning process, but by similar calculations between the terminology information extracted from Web texts and ontology categories. This results in a more accurate document classification since the meanings and relationships unique to each document are determined. Keywords: Document classification, Ontology, Web Page classification. [ABSTRACT FROM AUTHOR]

Raster map should be corrected after scanned because of the errors caused by paper map deformation. In the paper, the deficiency of the polynomial fitting method is analyzed. The paper introduces an ANN (Artificial Neural Network) correcting method that utilizes the advantage of its function approximation ability. In the paper, two types of ANNs, BP and GRNN, are designed for the correcting. The comparing experiment is done with the same data by the polynomial fitting and ANN methods, utilizing the MALAB. The experiment results show that the ANN methods, especially the GRNN method, performances far better than the polynomial fitting method does. [ABSTRACT FROM AUTHOR]

This paper investigates the fault detection and prediction of rhythmically soniferous products, such as clocks, watches and timers. Such products with fault cannot work steadily or probably cause malfunction. The authors extend the concept of computer audition and establish an architectural model of product fault prediction system based on probabilistic neural networks. The system listens to the product sound by the multimedia technology and the sound features are extracted to detect and predict faults by the neural network. The paper analyzes the reasons of timer faults and the corresponding sound features. Experiments are made in the laboratory to demonstrate the proposed method. The technology is expected to apply in factories in coming years for realizing automatic product test and improving efficiency of product inspection. [ABSTRACT FROM AUTHOR]

In this paper, selecting line segments as matching features in stereo vision, the orientation difference (O.D.) of the line segments is more deeply evaluated than the previous studies and two new constraints i.e. ordering and collinearity are proposed. The O.D. was supposed to be used for an indoor application, reducing candidates without elimination of the actual matches, especially if the line is horizontal or vertical with respect to the floor. The findings of this paper are as follows: 1) Applying a threshold on O.D. would result in a missing probability for the actual matches and this probability can be calculated for any given threshold, 2) The upper limit of O.D. for horizontal and vertical lines is a function of geometric parameters of the system, 3) An optimal tilt angle, whose results is the minimum upper limit for O.D., can be computed, 4) for disambiguation process, ordering and collinearity constraints are proposed. These are applied in a matching algorithm and their effectiveness is investigated on real stereo images. [ABSTRACT FROM AUTHOR]

In this paper, the line segment tracking is considered for an imaging system with translational motions (T.M.), rotational motions (R.M.) and arbitrary motions. Assuming a CCD camera with an arbitrary tilt angle installed on a mobile robot equipped with odometry system, to reduce the search space in the correspondence problem, two constraints are developed. These constraints are location and orientation differences (O.D.) for line segments in consecutive images. The findings of this paper include: 1)The development of the effect of camera tilt on the location constraint for T.M., 2)Illustrating that the upper bound of O.D. for both horizontal and vertical lines with respect to the floor, is a function of tilt in T.M., which can considerably be reduced and thus providing a tight constraint, 3)The development of the location constraint for R.M., 4) The development of the PDF and upper bound of O.D. for R.M., 5) The development of the location and O.D. constraint for arbitrary motions. Furthermore, the efficiency of these developed constraints in a line tracking algorithm was verified. [ABSTRACT FROM AUTHOR]

This paper presents a new description of image edge in the form of asymptote equation utilizing knowledge of differential geometry, and modifies a popular parlance in the traditional image processing. This paper develops the gradient concept and defines a generalized grads operator. The operator is a novel nonlinear transform, which inherits the strongpoint of noise suppression from the classical Gaussian differential filter and has a better edge extraction function. The experimental results show the proposed edge detection algorithm is powerful and effective. [ABSTRACT FROM AUTHOR]

The paper proposes a method to control the motion orbit of a cartoon object based on chaos theory. We first propose an adaptive delaying feedback controlling method which can control the Lorenz system to generate periodic, quasi-periodic and chaotic orbits. Then we apply our method to the topic of cartoon object motion controlling. It can enrich the motion type of the object thus enhances the visual effect of the cartoon video. The experiments validate the controlling results of the proposed method. Our research leads to a new area in which the chaotic theory is applied in the field of multimedia processing. [ABSTRACT FROM AUTHOR]

A Personal Assistant (PA) agent is a software agent capable of helping people to handle tasks in their workplace. The paper proposes a memory mechanism for personal assistant agents in order to enhance agent intelligence while working with the user or with other agents. Inspired by a case memory model in the domain of Case-Based Reasoning (CBR), this paper endows PA agents with a case memory mechanism, which results in improved PA agents: MemoPAs. We present the memory mechanism of MemoPA in detail, and report a first implementation of the method. Finally, future work is outlined for improving the memory mechanism. [ABSTRACT FROM AUTHOR]

A top-down experimental procedure for defect type recognition of epoxy-resin transformers by Partial Discharge (PD) is proposed. Most of the PD detection methods could be performed only at the shutdown period of equipments. By using Acoustic Emission (AE), the real-time and online detection could be reachable. Therefore, this paper conducted high voltage test of pre-faulty transformers and measured those PD signals for recognition needed. Afterward, the selected features that proposed in this paper can be extracted from these collected PD signals. According to these features, effective identification of their faulty types can be done using the proposed particle swarm optimization combined with neural network. Finally, with a view to apply in field, this research adds different noise levels to distort the original data. These distorted data are entered for subsequent testing. Research shows encouraging results that with 30% noise per discharge count, an 80% successful recognition rate can be achieved. [ABSTRACT FROM AUTHOR]

Electromagnetism-like algorithm (EM) is a population-based meta-heuristic which has been proposed to solve continuous problems effectively. In this paper, we present a new meta-heuristic that uses the EM methodology to solve the single machine scheduling problem. Single machine scheduling is a combinatorial optimization problem. Schedule representation for our problem is based on random keys. Because there is little research in solving the combinatorial optimization problem (COP) by EM, the paper attempts to employ the random-key concept enabling EM to solve COP in single machine scheduling problem. We present a hybrid algorithm that combines the EM methodology and genetic operators to obtain the best/optimal schedule for this single machine scheduling problem, which attempts to achieve convergence and diversity effect when they iteratively solve the problem. The objective in our problem is minimization of the sum of earliness and tardiness. This hybrid algorithm was tested on a set of standard test problems available in the literature. The computational results show that this hybrid algorithm performs better than the standard genetic algorithm. [ABSTRACT FROM AUTHOR]

Upon close examination of a set of industrial data from a large scale power plant, time varying behavior are discovered. If a fixed model is applied to monitor this process, false alarms will be inevitable. This paper suggests the use of adaptive models to cope with such situation. A recently proposed technique, fast algorithm for Moving Window Principal Component Analysis (MWPCA) was employed because of its following strength: (i) the ability in adapting process changes, (ii) the conceptual simplicity, and (iii) its computational efficiency. Its advances in fault detection is demonstrated in the paper by comparing with the conventional PCA. In addition, this paper proposed to plot the scaled variables in conjunction with MWPCA for fault diagnosis, which is proved to be effective in this application. [ABSTRACT FROM AUTHOR]

PVM (Parallel virtual machine) library is a tool which used processes large amounts of data sets. This paper wants to achieve a high performance solution that exploits PVM library and parallel computers to solve ICA (Independent Component Analysis) problem. The paper presents parallel power ICA implementations to decomposition data sets. Power iteration (PI) is an algorithm for independent component analysis, which has some desired features. It has higher performance and data capacity than current sequential implementations. This paper, we show the power iteration algorithm which learning updating is in the form of matrix transformation . From power iteration algorithm, we develop parallel power iteration algorithm and implement parallel component decomposition solution. At last, experimental results, analysis and future plans are presented. [ABSTRACT FROM AUTHOR]

In this paper, global exponential stability of Cohen-Grossberg neural networks with reaction-diffusion and Dirichlet boundary conditions is considered by using an approach based on the delay differential inequality and the fixed-point theorem. Some sufficient conditions are obtained to guarantee that the reaction-diffusion Cohen-Grossberg neural networks are globally exponentially stable. The results presented in this paper are the improvement and extension of the existed ones in some existing works. [ABSTRACT FROM AUTHOR]

This paper analyzes global convergence and learning parameters of the back-propagation algorithm for quadratic functions. Some global convergence conditions of the steepest descent algorithm are obtained by directly analyzing the exact momentum equations for quadratic cost functions. In addition, in order to guarantee the convergence for a given learning task, the method is obtained to choose the proper learning parameters. The results presented in this paper are the improvement and extension of the existed ones in some existing works. [ABSTRACT FROM AUTHOR]

The automatic relevance determination (ARD) shows good performance in many applications. Recently, it has been applied to brain current estimation with the variational method. Although people who use the ARD tend to pay attention to one benefit of the ARD, sparsity, we, in this paper, focus on another benefit, generalization. In this paper, we clarify the generalization error of the ARD in the case that a class of prior distributions is used, and show that good generalization is caused by singularities of the ARD. Sparsity is not observed in that case, however, the mechanism that the singularities provide good generalization implies the mechanism that they also provide sparsity. [ABSTRACT FROM AUTHOR]

This paper presents iris recognition for personal identification using neural networks. Iris recognition system consists of localization of the iris region and generation of data set of iris images and then iris pattern recognition. One of the problems in iris recognition is fast and accurate localization of the iris image. In this paper, fast algorithm is used for the localization of the inner and outer boundaries of the iris region. Located iris is extracted from an eye image, and, after normalization and enhancement it is represented by a data set. Using this data set a neural network is applied for the classification of iris patterns. Results of simulations illustrate the effectiveness of the neural system in personal identification. [ABSTRACT FROM AUTHOR]

The www is a dynamic environment and it is difficult to capture the user preferences and interests without interfering with the normal activity of the user. Hence to improve user searching in the World Wide Web, we have proposed a personalized search system that supports user searches by learning about user preferences and by observing responses to prior search experiences aided by a new index called the User Conceptual Index (UCI). This paper models every user's search behavior as a User's Search Behavior (USB) Graph. The main focus of this paper is the analysis of the USB graph and the redesign of the UCI using the results arrived from the analysis. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel approach for facial expression analysis and recognition. The main contributions of the paper are as follows. First, we propose an efficient facial expression recognition scheme based on the detection of keyframes in videos where the recognition is performed using a temporal classifier. Second, we use the proposed method for extending the human-machine interaction functionality of the AIBO robot. More precisely, the robot is displaying an emotional state in response to the recognized user's facial expression. Experiments using unseen videos demonstrated the effectiveness of the developed method. [ABSTRACT FROM AUTHOR]

The aim of this paper is to introduce a novel, biologically inspired approach to extract visual features relevant for controlling and understanding reach-to-grasp actions. One of the most relevant of such features has been found to be the grip-size defined as the index finger-tip - thumb-tip distance. For this reason, in this paper we focus on this feature. The human visual system is naturally able to recognize many hand configurations - e.g. gestures or different types of grasps - without being affected substantially by the (observer) viewpoint. The proposed computational model preserves this nice ability. It is very likely that this ability may play a crucial role in action understanding within primates (and thus human beings). More specifically, a family of neurons in macaque's ventral premotor area F5 have been discovered which are highly active in correlation with a series of grasp-like movements. This findings triggered a fierce debate about imitation and learning, and inspired several computational models among which the most detailed is due to Oztop and Arbib (MNS model). As a variant of the MNS model, in a previous paper, we proposed the MEP model which relies on an expected perception mechanism. However, both models assume the existence of a mechanism to extract visual features in a viewpoint independent way but neither of them faces the problem of how this mechanism can be achieved in a biologically plausible way. In this paper we propose a neural network model for the extraction of visual features in a viewpoint independent manner, which is based on the work by Poggio and Riesenhuber. [ABSTRACT FROM AUTHOR]

The aim of this paper is to compare the effectiveness of various computational intelligence approaches applied to the task of retrieving musical rhythm from musical symbolic files. The study presented in this paper describes how Artificial Neural Networks and Rough Sets can be used for searching the metric structure of musical files. The described approaches are based on examining physical attributes of sound that are most significant in determining the placement of a particular sound in the accented location of a musical piece. The results of the experiments show that the approach based solely on duration is sufficient enough to retrieve the metric structure of rhythm from musical files. [ABSTRACT FROM AUTHOR]

This paper is an extension of our earlier work in combining strengths of rough set theory and neuro-fuzzy decision trees in classifying software defect data. The extension includes the application of a rough-fuzzy classification trees to classifying defects. We compare classification results for five methods: rough sets, neuro-fuzzy decision trees, partial decision trees, rough-neuro-fuzzy decision trees and rough-fuzzy classification trees. The analysis of the results include a paired t-test for accuracy and number of rules. The results demonstrate that there is improvement in classification accuracy with the rough fuzzy classification trees with a minimal set of rules. The contribution of this paper is a comparative study of several hybrid approaches in classifying software defect data. [ABSTRACT FROM AUTHOR]

Currently used data-mining algorithms treat data globally. Nevertheless, with such methods, potentially useful knowledge that relates to local phenomena may be undetected. In this paper, we introduce new patterns in a form of local frequent events and epochs, boundaries of which correspond to discovered changes in a data stream. A local frequent event is an event which occurs in some period of time frequently, but not necessarily in the whole data stream. Such an event will be called a frequent event in a data stream. An epoch is understood as a sufficiently large group of frequent events that occur in a similar part of the data stream. The epochs are defined in such a way that they do not overlap are separated by so called change periods. In the paper, we discuss some potential applications of the proposed knowledge. Preliminary experiments are described as well. [ABSTRACT FROM AUTHOR]

In T −fuzzy rough sets a fuzzy T −similarity relation is employed to describe the similar degree between two objects and to construct lower and upper approximations for arbitrary fuzzy sets. The existing researches on T −fuzzy rough sets mainly concentrate on constructive and axiomatic approaches of lower and upper approximation operators. In this paper we define attribute reduction based on T −fuzzy rough sets. The structure of proposed attribute reduction is investigated in detail by the approach of discernibility matrix. At last an example is proposed to illustrate our idea in this paper. [ABSTRACT FROM AUTHOR]

The subject-matter of the consideration touches the problem of vagueness. The notion of the rough set, originated by Zdzisław Pawlak, was constructed under the influence of vague information and methods of shaping systems of notions leading to conceptualization and representation of vague knowledge, so also systems of their scopes as some vague sets. This paper outlines some direction of searching for a solution to this problem. In the paper, in connection to the notion of the rough set, the notion of a vague set is introduced. Some operations on these sets and their properties are discussed. The considerations intend to take into account a classical approach to reasoning, based on vague premises, and suggest finding a logic of vague sentences as a non-classical logic in which all counterparts of tautologies of classical logic are laws. [ABSTRACT FROM AUTHOR]

An optimization algorithm for dynamic Bayesian networks (DBN) based on Bayesian optimization algorithm (BOA) is developed for learning and constructing the DBN structure. In this paper, we first introduce some basic theories and concepts of probability model evolutionary algorithm. Then we describe, the basic mode for constructing DBN diagram and the mechanism of DBN structure learning based on BOA. The DBN structure learning based on BOA consists of two parts. The first part is to obtain the structure and parameters of DBN in terms of a good solution, and the second part is to produce new groups according to the obtained DBN structure. In this paper, the DBN learning is achieved by genetics algorithm based on a greedy mechanism. The DBN inference is performed by a forward-simulation algorithm. Simulation results are provided to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Generalization problem is a key problem in NN society, which can be grouped into two classes: the generalization problem with unlimited size of training sample and that with limited size of training sample. The generalization problem with limited size of training sample is considered in this paper. Similar to margin maximization criterion in SVM, we propose a margin maximization training algorithm for BP network to further improve the generalization ability of BP network. Experimental results show that the margin maximization training algorithm proposed in this paper does improve the performance of BP network, and shows a comparable performance with SVM. [ABSTRACT FROM AUTHOR]

Chaotic noise cancellation has potential application in both secret communication and radar target identification. To solve the problem of parameter estimation in chaotic noise, a novel radial basis function neural network (RBF-NN) -based chaotic time series data modeling method is presented in this paper. Together with the spectral analysis technique, the algorithm combines neural network's ability to approximate any nonlinear function. Based on the flexibility of RBF-NN predictor and classical amplitude spectral analysis technique, this paper proposes a new algorithm for parameter estimation in chaotic noise. Analysis of the proposed algorithm's principle and simulation experiments results are given out, which show the effective of the proposed method. We conclude that the study has potential application in various fields as in secret communication for narrow band interference rejection or attenuation and in radar signal processing for weak target detection and identification in sea clutter. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel image digital watermarking technique based on Kernel Independent Component Analysis (KICA). Use the nice characteristic of the KICA, which can results the blind separation of nonlinearly mixed signals, the imperceptibility and robustness requirements of watermarks are fulfilled and optimized. In the proposed scheme, the watermark image is first transformed by Arnold method, and then embedded into the lowest frequency subband in DWT domain. The recovery of owner's image is turning the watermarked image into DWT domains then use KICA to extract the watermark. Finally the watermark is transformed by Arnold method again, so we can get the original watermark image. Experimental results show that the proposed method gains better performance in robustness than that of ICA with respect to traditional image processing including cropping, filtering, add noise and JPEG image compression. [ABSTRACT FROM AUTHOR]