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 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 face images are varying environment database from external illumination. Therefore we propose parallel cascade face detector. We define context image illumination and distinguish context using unsupervised learning. Many unsupervised method is available to distinguish varying illuminant images. This approach can be distribution face image and we can make the classifier for face image context. Therefore, in this paper, we parallel classifier that is strutted cascade classifier of two methods. In first classifier, we use sub-sampling feature extraction and in second classifier we use wavelet transformation method. We achieved very encouraging experimental results. Our method is enhancement varying illumination environment. [ABSTRACT FROM AUTHOR]

This paper summarizes a new concept to determine principal curves for nonlinear principal component analysis (PCA). The concept is explained within the framework of the Hastie and Stuetzle algorithm and utilizes spline functions. The paper proposes a new algorithm and shows that it provides an efficient method to extract underlying information from measured data. The new method is geometrically simple and computationally expedient, as the number of unknown parameters increases linearly with the analyzed variable set. The utility of the algorithm is exemplified in two examples. [ABSTRACT FROM AUTHOR]

In this paper, we have proposed a fast affine transform method for real-time machine vision applications. Inspection of parts by machine vision requires accurate, fast, reliable, and consistent operations, where the transform of visual images plays an important role. Image transform is generally expensive in computation for real-time applications. For example, a transform including rotation and scaling would require four multiplications and four additions per pixel, which is going to be a great burden to process a large image. Our proposed method reduces the complexity substantially by removing four multiplications per pixel, which exploits the relationship between two neighboring pixels. In addition, this paper shows that the affine transform can be performed by fixed point operations with marginal error. Two interpolation methods are also tried on top of the proposed method in order to test the feasibility of fixed point operations. Experimental results indicated that the proposed algorithm was about six times faster than conventional ones without any interpolation and five times faster with bilinear interpolation. [ABSTRACT FROM AUTHOR]

The Although a variety of standardization and Learning Management System (LMS) were produced to develop and effectively manage web contents in response to active diffusion of internet application, practical changes to assist online learners are not yet to be found. The core purpose of the LMS system proposed in this paper is to help learners pursue proactive and self-oriented education by allowing learners to proactively configure their own content, that is, learners no longer have to be restricted by prescribed sequence of lectures. In this paper, I would like to introduce a LMS that can support self-leading education by providing various types of learners at Virtual University with delicately organized educational contents for maximum efficiency. The system allows a learner to select a lecture or a chapter which has been presorted to meet his educational needs and intellectual ability. [ABSTRACT FROM AUTHOR]

This paper considers the capacities of the elements in a set E efficiently so that the total cost for the increment of capacity can be decrease to maximum extent while the final expansion capacity of a given family F of subsets of E is with a given limit bound. The paper supposes the cost w is a fuzzy variable. Network bottleneck capacity expansion problem with fuzzy cost is originally formulated as Expected value model according to some criteria. For solving the fuzzy model efficiently, network bottleneck capacity algorithm, fuzzy simulation, neural network(NN) and genetic algorithm(GA) are integrated to produce a hybrid intelligent algorithm. [ABSTRACT FROM AUTHOR]

This paper investigates the global exponential stability of Takagi-Sugeno Fuzzy cellular neural networks with time-varying delays. Using the reduction to absurdity, a less conservative delay-independent stability criterion is derived to guarantee the exponential stability of Takagi-Sugeno Fuzzy cellular neural networks with time-varying delays. Since our model is more general than some existing works, the results presented in this paper are the improvement and extension of the existed ones. [ABSTRACT FROM AUTHOR]

A modified least squares support vector machines (LS-SVM) approach, which treats the training data points differently according to their different degrees of importance, is proposed in this paper. On each data point, a support vector degree is defined and it is associated with the corresponding absolute value of Lagrange multiplier. The experiment of identification of pH neutralization process with polluted measuring data is shown in this paper and the result indicates that the method is effective in identification of nonlinear system. By contrast with the basic LS-SVM, the result also shows the priority of the presented new algorithm. [ABSTRACT FROM AUTHOR]

In the wireless sensor networks, sensor nodes are generally clustered under the necessity. GPSR which Brad Karp et al. proposed, is routing algorithm that has follow advantages: GPSR is localized algorithm to find routes and sensor nodes have few data for routing. These features make GPSR be adapted to sensor networks. But, forwarding a packet is able to be failed because of geographic data error or mobility of sensor nodes. In order to improve such a problem, this paper proposes clustered GPSR by fuzzy membership. [ABSTRACT FROM AUTHOR]

This paper presents a new interest point descriptors representation method based on independent components analysis (ICA). The aim of this algorithm is to find a meaningful image subspace and more compact descriptors. Combination the descriptors with an effective interest point detector, the proposed algorithm has a more accurate matching rate besides the robustness towards image deformations. The proposed algorithm first finds the characteristic scale and the location for the interest points using Harris-Laplacian interest point detector. We use Haar wavelet transform on the neighborhood of the interest points and get low frequency gradient feature vectors. Then ICA is used to model the subspace and reduces the dimension of the feature vectors. The experiments show the efficiency of the proposed algorithm. [ABSTRACT FROM AUTHOR]

In this paper, we propose a system to categorize audio in 7 classes. For classification features, we use the mean and variance of RMS, ZCR, fundamental frequency and frequency peak which are extracted from every frame of 25ms length. In addition to the audio content classification, we also perform speaker identification with the voice sequences extracted automatically using our proposed method. The accuracy of our proposed scheme reaches 93.8% in categorizing audio signal and 80% in the speaker identification process. [ABSTRACT FROM AUTHOR]

OVAL (Open Vulnerability Assessment Language) is a standard language which is used to detect the vulnerability of local system based on the system characteristics and configurations. It is suggested by MITRE. OVAL consists of XML schema and SQL query statements. XML schema defines the vulnarable points and SQL query detects the vulnerable and weak points. This paper designed and implemented the vulnerability assesment tool with OVAL to detect the weak points in security countermeasure block. It has more readability, reliability, scalability and simplicity than traditional tools. [ABSTRACT FROM AUTHOR]

In many complicated constrained optimization problems, intelligent searching technique based algorithms are very inefficient even to get a feasible solution. This paper presents an enhanced heuristic searching algorithm to solve this kind of problems. The proposed algorithm uses known feasible solutions as heuristic information, then orients and shrinks the search spaces towards the feasible set. It is capable of improving the search performance significantly without any complicated and specialized operators. Benchmark problems are tested to validate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Management system is based on agent is divided into two parts such as operational and collaborative customer management and analytical customer management for ubiquitous environment. This system is in charge of marketing automation and customer services which are included in operational and collaborative is developed. And the analytical engine part which can create report in using and analyzing data mining technology which is based on the rules with collecting customer information is developed, too. The operational customer management part takes part in various channels and customer check and it is consisted of many components which are necessary for communication. This was implemented through basis of specialized components and it supports the concept of ASP service. Through the development of a module, which helps to use existing data environment more effectively, it supports and creates an environment that users can use easily. [ABSTRACT FROM AUTHOR]

This paper presents a watermarking algorithm in the discrete wavelet transform domain using evolutionary algorithm. The proposed algorithm consists of wavelet-domain watermark insertion and genetic algorithm-based watermark extraction. More specifically watermark is inserted to the low-frequency region of wavelet transform domain, and watermark extraction is efficiently performed by using the evolutionary algorithm. The proposed watermarking algorithm is robust against various attacks such as JPEG image compression and geometric transformations. [ABSTRACT FROM AUTHOR]

This paper proposed a new kind of immune genetic algorithm (IGA) according to the current algorithms solving the reactive power optimization. The hybrid algorithm is applied in reactive power optimization of power system. Adaptive crossover and adaptive mutation are used according to the fitness of individual. The substitution of individuals is implemented and the multiform of the population is kept to avoid falling into local optimum. The decimal integer encoding and reserving the elitist are used to improve the accuracy and computation speed. The flow chart of improved algorithm is presented and the parameter of the immune genetic algorithm is provided. The procedures of IGA algorithm are designed. A standard test system of IEEE 30-bus has been used to test. The results show that the improved algorithm in the paper is more feasible and effective than current known algorithms. [ABSTRACT FROM AUTHOR]

When processing an image of multiple occluded humans, segmenting them is a prerequisite for higher-level tasks such as tracking and activity analysis. Although a human observer can easily segment target humans partly occluded among themselves in an image, automatic segmentation in computer vision is difficult. In this paper, the use of a probabilistic neural network is proposed to learn various outline shape patterns of a foreground image blob of occluded humans, and then to segment the blob into its constituents. The segmentation is here regarded as a two-class pattern recognition problem; segmentable positions constitute a class and other positions constitute the other. The technique proposed is useful particularly for low-resolution images where existing image analysis techniques are difficult to be applied. [ABSTRACT FROM AUTHOR]

The simulation line is usually used to imitate the frequency characteristic of a real long transmission line. This paper proposes a novel design scheme of simulation line using back propagation neural network (BP NN). A BP NN is trained to correspond with the line's transfer function and then implemented by field programmable gate array (FPGA) for application in real time. The activation function of NN is approximated with a high-speed symmetric table addition method (STAM), which reduces the amount of memory required. For an underwater coaxial cable that is 10000m long, a simulation line is hardware realized and has been successfully used in the study of digital image transmission. [ABSTRACT FROM AUTHOR]

In this paper, we present experiments comparing different training algorithms for Radial Basis Functions (RBF) neural networks. In particular we compare the classical training which consists of an unsupervised training of centers followed by a supervised training of the weights at the output, with the full supervised training by gradient descent proposed recently in same papers. We conclude that a fully supervised training performs generally better. We also compare Batch training with Online training and we conclude that Online training suppose a reduction in the number of iterations. [ABSTRACT FROM AUTHOR]

The adaptive control method based on orthogonal function neural network is proposed for a class of chaotic nonlinear systems. The adaptive controller is constructed by using a single hidden layer Chebyshev orthogonal function neural network which has advantages such as simple calculation and fast convergence. The adaptive learning law of orthogonal neural network is derived to guarantee that the adaptive weight errors and the tracking error are bounded from Lyapunov stability theory. The uncertain nonlinear system with the external disturbances can track the desired reference trajectory with bounded errors by means of the adaptive feedback controller. Based on the orthogonal function neural network, the control of chaotic systems with uncertainties is studied. The results show that the approach proposed in this paper can overcome effectively the external disturbances. [ABSTRACT FROM AUTHOR]

Color appearance of the same object between different viewpoints may be different due to inconsistent illumination or inevitable errors and noises resulting from imaging devices, which brings difficulties to the subsequent multi-view image compression and arbitrary viewpoint rendering. Thus, color correction is necessary for multi-view applications. In this paper, a color correction method of multi-view images is proposed based on region correspondence. Through expectation-maximum segmentation for reference image and current viewpoint image, mapping relationship is built with similar statistical model between two regions in the two images. Experimental results show that the proposed method can be used to correct the color of viewpoint images according to the color of the reference image, and preserve the visual impression of the viewpoint images simultaneously. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new arbitrary view image generation method using panoramic-based image morphing for large-scale scenes. The panoramic images are sampled with large step size. From these panoramic images, intermediate panoramic images are generated by bi-directional image morphing and finally arbitrary view images are generated by light field rendering. It is shown that an arbitrary view image with reasonably high quality is obtained using a few captured panoramic images. [ABSTRACT FROM AUTHOR]

This paper describes a shape recovery framework based on scaled orthographic factorization and incremental 3D point cloud registration. While the camera is moving freely in unknown environments, point features are automatically extracted and tracked. Using the tracked features in three consecutive frames, 3D locations are recovered and partial point subsets are immediately registered to the global point cloud. The advantage of the proposed incremental method is that the registration process does not rely on past frames for estimating the global 3D point cloud. The shape recovery and registration steps are combined as a single procedure that processes each input frame only once. All the internal steps are linear and non-iterative, hence the procedure is applicable to on-line interactive applications. Experimental results showing the feasibility of the presented approach are also provided. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new multi-modal feature integration for secure authentication. We introduce behavioral information as well as biometrics information for the person of interest to test his verification. For continuous authentication, temporal score integration method is presented that incorporates biometrics and behavioral features. The proposed method was evaluated under several real situations and promising results were obtained. [ABSTRACT FROM AUTHOR]

Non-intrusive methods for eye tracking are important for many applications of vision-based human computer interaction, such as driver fatigue detection, eye gaze replacing the hand operating mouse, eye typing instead of manually depressing keys as a virtual keyboard, eye gaze correction for video conferencing, interactive assistant application for disabled users, etc. However, due to the eye motion be the high nonlinearity, the obstacles of robustness of external interference and accuracy of eye tracking, these tend to significantly limit their scope of application. In this paper, we present a strong tracking finite-difference extended Kalman filter algorithm, and overcome the modeling of nonlinear eye tracking. In filtering calculation, strong tracking factor is introduced to modify prior covariance matrix to improve the accuracy of the filter. The filter uses finite-difference method to calculate partial derivatives of nonlinear functions to eye tracking. The last experimental results show validity of our method for eye tracking under realistic conditions. [ABSTRACT FROM AUTHOR]

In this paper, the new STTD receiver for multimedia communication system in ubiquitous network environment is proposed and the performance of them is analyzed. For the multimedia systems that equip the transmit diversity for forward channel, it is well known that the Tx diversity technique is appropriate for the high rate communications. Adopting the proposed STTD receiver, at 10-2 ~ 10-4 in bit error rate, the SNR of 0.5dB ~ 2dB performance improvement in AWGN and Rayleigh fading environment has been achieved. [ABSTRACT FROM AUTHOR]

Presently, the advanced computing capabilities of information communication hardware and equipment far exceed the ability of software that utilizes these features. Mobile phones have extraordinary features for multimedia and communications and are the prime example of current software limitations. The multimedia usage of mobile phones is limited in its speed because of the computations requirements for internet connection, poor user interface and battery power drainage, due to heavy computation requirements. While these devices are very convenient and portable, there are serious problems for users to fully utilize the data processing and communications capabilities. This paper proposes an improvement of mobile phone service by developing software based on an X-internet approach. [ABSTRACT FROM AUTHOR]

This paper proposed a new method of extracting texture features based on contourlet domain in RGB color space. In addition, the application of these features for bark classification applying radial basis probabilistic network (RBPNN) has been introduced. In this method, the bark texture feature is firstly extracted by decomposing an image into 6 subbands using the 7-9 biorthogonal Debauches wavelet transform, where each subband is fed to the directional filter banks stage with 32 directions at the finest level, then the mean and standard deviation of the image output are computed. The obtained feature vectors are fed up into RBPNN for classification. Experimental results show that, features extracted using the proposed approach can be more efficient for bark texture classification than gray bark image. [ABSTRACT FROM AUTHOR]

Benchmarking is a technique that engages and executes a series of measures to change indexes of the Decision Making Unit (DMU) to excellent by using the gap analysis information between the DMU and benchmark. In this paper, a DEA-Benchmarking model based on the theory of maximum entropy is proposed and the conception of Entropy-DEA efficiency is defined. According to the optimization model based on the theory of maximum entropy, the Entropy-DEA efficient DMUs is regarded as benchmarks, which have more advantages and direction than DEA efficient DMUs. The measure method and existence property of Entropy-DEA efficiency are all analyzed in this paper. [ABSTRACT FROM AUTHOR]

Ultra-wideband (UWB) technology has been considered to offer an innovative solution for future short-range high-speed wireless communication. Interference suppression is important for the UWB devices to operate over spectrum with established narrowband systems. But it's difficult to secure a QoS of high quality because it coexists other system that was used same band. We need a adaptable change of the system that follow the changing channel environment. In this paper, derived monocycle pulse that application is possible within permission frequency band that establish in The Federal Communications Commission(FCC). Also, PPM modulated TH IR system simulator is designed in Two-path environment. It is analyzed monocycle pulse specific property and system performance using simulator. [ABSTRACT FROM AUTHOR]

Multipath fading causes decrease of system performance in wireless communication environment. Therefore, intelligent high-speed communication system which can effectively reduce ever-changing multipath interference is necessary in order to provide high quality data service. In this paper, we analyzed channel performance of intelligent UWB communication system in indoor radio fading environment that consider amplitude characteristic of channel. Also, we improved performance of system that it is decreased in fading environment employing rake receiver and convolution encoding techniques. [ABSTRACT FROM AUTHOR]

An adaptable designated group signature is presented in this paper. In contrast with traditional group oriented signature, the new one laid a strong emphasis on how to improve the signer's efficiency. In fact, this new type of designated group signature can be seen as a type of designated verifier signature. In contrast with the ordinary designated verifier signature, it does not designate one member but several members to independently verify the signature. The designated members, who can independently verify the signature, come into a group. This scheme can ensure the anonymity of the verifiers. This type of signature can be used in such system that the compute resource is limited, such as the broadcast protocols of the mobile telephone in the mobile networks. [ABSTRACT FROM AUTHOR]

Home and abroad, the M/W intermediary service bandwidth is between 3 and 30 GHz. However, recently the bandwidth is getting insufficient due to the increasing number of services and the capacities. Thus FPU (Field Pick UP) Transmission System, the mobile broadcasting intermediary link, needs to be evaluated in the system capacity by means of simulation in order to estimate whether or not the number of channels and the places are changeable when the bandwidth is rearranged in the future. In this paper, we have modeled of Digital FPU link system in M/W band portable digital wireless transmission system. Also, we analyzed digital FPU system performance adopt adaptive modulation scheme in variable channel environment. [ABSTRACT FROM AUTHOR]

In this paper, we propose an intelligent middleware framework based on the concept of the autonomic computing. Because decisions for optimizing a system are highly dependent on the status of the system in the near future, predictions of the status can boost up the performance of system. Bayesian networks that are frequently used in the domain of user modeling are adopted as a diagnosis engine. Since there are many different kinds of behavioral patterns, it is impossible to model all of them into a single model. A Bayesian network is learned for each typical behavioral pattern. Experimental results show that the proposed method offers a promising way of intelligent middlewares. [ABSTRACT FROM AUTHOR]

Object detection is very important to service robots. Many tasks for service such as delivery, cleaning, and health-care for elderly people are strongly related to objects. Conventional approaches for object detection are mainly based on the geometric models, because they have been applied to static environments. In indoor environments having uncertainty, they have limitation in some situations where interesting objects are occluded by other ones or small in the scene. Context information can be helpful to overcome these uncertain situations. In this paper, we adopt objects as context information to allow for service robots to predict the probability of interesting objects through observed ones. For this, an object relationship model based on Bayesian network (BN) and integration method are proposed. Experimental results confirm that the proposed method predicts the objects very well. [ABSTRACT FROM AUTHOR]

Aircraft Maintenance Review Board Report (MRB Report) is necessary one of the most important Continuing Airworthiness documents. Determination Maintenance interval in MRB Report depends mainly on experience and there isn't a rigorous and quick method. The paper proposes a multi-stage framework for determination maintenance interval using case-based reasoning (CBR), which uses fuzzy generalized nearest-neighbor matching (FNN) to retrieve case and fuzzy Group decision making to determine attributes' weight. An example illustrates the proposed approach and the average relative error is under 3.21%. Finally the developed civil aircraft CBR-MRBR decision support system is described. The method enriches the MRB Report developing method. [ABSTRACT FROM AUTHOR]

Airline companies usually implement revenue management to increase profits. The revenue management can be performed through seat inventory management. The current system to book seats is first-come first-served. This approach tends to sell low-price seats because low-price requests often appear earlier. It also results in low revenue. In this paper, an expected dynamic probability method and a case-based seat allocation system are proposed to enhance the performance of the seat inventory management. Extensive studies are conducted to compare the performance of first-come first-served method, expected dynamic probability method, and case-based decision support system. The result indicates that the case-based seat allocation system outperforms the other methods. [ABSTRACT FROM AUTHOR]

In order to discover and manage enterprise businesses effectively, enterprise decision makers need to understand enterprise business processes from various aspects using sophisticated process simulation and analysis tools. Based on ABM technology (Activity-Based Management), this paper presents intelligent business process analysis concepts and metric measurement models for six kinds of process flows within manufacturing enterprises: activity flow, product flow, resource flow, cost flow, cash flow, and profit flow. The proposed process flow analysis approach has been developed as a dynamic enterprise process analysis tool within a process simulation and optimization environment. [ABSTRACT FROM AUTHOR]

This paper presents a Sudoku solver based on the energydriven neural-network (NN) model, called the Q'tron NN model. The rules to solve Sudoku are formulated as an energy function in the same form as a Q'tron NN's. The Q'tron NN for Sudoku can then be built simply by mapping. Equipping the NN with the proposed noise-injection mechanism, the Sudoku NN is ensured local-minima free. Besides solving Sudoku puzzles, the NN can also be used to generate Sudoku puzzles. [ABSTRACT FROM AUTHOR]

High speed networks with high delays present a unique environment where TCP can not efficiently utilize the full bandwidth. In this paper, we present a simple and novel Time Based Congestion Control scheme (TBCC for short) that alleviates RTT unfairness while supporting TCP friendliness and bandwidth scalability. We provide the general framework for TBCC protocol modifications. Then, using a simplified design, we illustrate the scalability of TBCC for improving the efficiency, without sacrificing the fairness of TCP. We assess the RTT unfairness of TBCC and show that the RTT unfairness of TBCC can be firmly guaranteed to be no worse than that of unmodified TCP. Compared with similar protocols such as HSTCP and STCP, TBCC can guarantee the bandwidth scalability and achieve better fairness. The flexible framework of TBCC provides a whole class of design options for improving the performance of TCP in high speed networks. [ABSTRACT FROM AUTHOR]

This paper discusses the single-machine large-scale rescheduling problem with efficiency and stability as criteria, where more disruptions arise during the execution of schedule. A rolling strategy is driven by disruption events and the reactive partial rescheduling (PR) is adopted at each disruption. Two types of PR sub-problem are designed respectively for the procedural and the terminal PR-horizons. A local search algorithm is used to solve the PR sub-problems. Computational results show that the rolling PR can greatly improve the schedule stability with little sacrifice in schedule efficiency and consistently outperforms the rolling RSR. [ABSTRACT FROM AUTHOR]

Optimum multiuser detection (OMD) in direct-sequence code-division multiple access (DS-CDMA) communication systems is an NP-complete combinatorial optimization problem. The first contribution of this paper is the theoretical investigation of the OMD problem. Its fitness landscape is specified by a set of neighborhoods of all points of the search space. The number and the distributions of local optima are studied in detail. Investigation results give hints how to choose the modification operators and design more efficient random search heuristics for this problem. Then an efficient iterated local search algorithm is proposed for multiuser detection and simulation results show that it can provide rather good performance for cases where other algorithms perform poorly. [ABSTRACT FROM AUTHOR]

The goal of this paper is to solve an optimal consumption-investment problem with fuzzy financial coefficients. A multi-objective fuzzy decision-making model for consumption-investment problem is proposed, based on the uncertainty in some economic factors, to maximize the consumption utility and to maximize the total profit in investment and to minimize the risk in investment. The fuzzy optimal consumption-investment strategy is characterized by maximizing the satisfactory grade of the decision-maker. Finally an effective algorithm is proposed to solve the problem and a numerical example shows the effectiveness and feasibility of this method. [ABSTRACT FROM AUTHOR]

The relaxation algorithm for linear programming is revised in this paper. Based on cluster structure, a parallel revised algorithm is presented. Its performance is analyzed. The experimental results on DAWNING 3000 are also given. Theoretical analysis and experimental results show that the revised relaxation algorithm improves the performance of the relaxation algorithm, and it has good parallelism and is very robust. Therefore, it can expect to be applied to the solution of the large-scale linear programming problems rising from practical application. [ABSTRACT FROM AUTHOR]

As a novel evolutionary computing technique, recently Differential Evolution (DE) has attracted much attention and wide applications due to its simple concept and easy implementation. However, all the control parameters of the classic DE (crossover rate, scaling factor, and population size) keep fixed during the searching process. To improve the performance of DE, an improved DE (IDE) with dynamic population size is proposed in this paper. Simulation results and comparisons based on some well-known benchmarks and an IIR design problem show the good efficiency of the proposed IDE. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method of classifier selection for efficient object recognition based on evolutionary computation and data context knowledge called Evolvable Classifier Selection. The proposed method tries to distinguish the data characteristics of input image (data contexts) and selects a classifier system accordingly using the genetic algorithm. It stores its experiences in terms of the data context category and the artificial chromosome so that the context knowledge can be accumulated and used later. The proposed method operates in two modes: the evolution mode and the action mode. It can improve its performance incrementally using GA in the evolution mode. Once sufficient context knowledge is accumulated, the method can operate in real-time. The proposed method has been evaluated in the area of face recognition. Data context-awareness, modeling and identification of input data as data context categories, is carried out using SOM. [ABSTRACT FROM AUTHOR]

Recently, many new flow control mechanisms derived from classic Kelly model are proposed to solve network congestion problem. They perform well in stability, fairness or robustness. However, most of their convergence rates are linear since in classic Kelly model, link price is only positive. In addition, some need to introduce extra packet header to get price information. This paper presents a novel flow control model based on Kelly model in which the link price can be negative to improve the convergence rate. Further, The proposed model uses two bits of ECN field to feed back price instead of introducing new packet header data. Thus it can implement flow control scheme achieving exponential convergence in traditional TCP/IP datagram format. NS2 simulation results show that our model can keep fairness and asymptotic stability with more rapid convergence rate. [ABSTRACT FROM AUTHOR]

Image enhancement with various image filters might be better than that with a single filter, but it is very difficult to find a set of appropriate filters. In this paper, we propose a novel method that uses the genetic algorithm to find a filter set for reducing impulsive noises in images. The proposed method has been verified with benchmark images in image enhancement. Moreover, it does not require any expert knowledge to find the type and order of filters for a given domain, so it can be easily applied to other applications of image processing. [ABSTRACT FROM AUTHOR]

In modern production industries, computer aided systems have been improving the efficiency and convenience of the various stages of work. However, as the complexity of computerized production systems increases, various techniques are still necessary. The problem we addressed occurs in computer systems that automatically make manufacturing process plans in the metal grating manufacturing industry. In the system, the merging of tasks as a work unit is important to reduce the material loss. However, there is no guarantee that merging always reduces the material loss. So, operators must compare the material loss rates of diverse merging cases to find a near-optimal solution that provides a low material loss rate. In this paper, we apply genetic algorithms to search the near-optimal solution of a planning problem focused on the reduction of material loss. In order to reflect the domain dependent characteristics, we apply genetic algorithms in two levels related each other. [ABSTRACT FROM AUTHOR]