Repacking is an efficient scheme for bandwidth packing problem (BPP) in centralized networks (CNs), where a central unit allocates bandwidth to the rounding terminals. In this paper, we study its performance by proposing a new formulation of the BPP in the CN, and introducing repacking scheme into next fit algorithm in terms of the online constraint. For the realistic applications, the effect of call demand distribution is also exploited by means of simulation. The results show that the repacking efficiency is significant (e.g. the minimal improvement about 13% over uniform distribution), especially in the scenarios where the small call demands dominate the network.

We propose a recursive clustering and order restriction (R-CORE) method for learning large-scale Bayesian networks. The proposed method considers a reduced search space for directed acyclic graph (DAG) structures in scoring-based Bayesian network learning. The candidate DAG structures are restricted by clustering variables and determining the intercluster directionality. The proposed method considers cycles on only c_max(«n) variables rather than on all n variables for DAG structures. The R-CORE method could be a useful tool in very large problems where only a very small amount of training data is available.

A communication means is presented for patients with amyotrophic lateral sclerosis in totally locked-in state who are completely unable to move any part of the body and have no usual communication means. The method utilizes changes in cerebral blood volume accompanied with changes in brain activity. When a patient is asked a question and the answer to it is 'yes', the patient makes his or her brain active. The change in blood volume at the frontal lobe is detected with near-infrared light. The instantaneous amplitude and phase of the change are calculated, and the maximum amplitude and phase change are obtained. The answer 'yes' or 'no' of the patient is detected using a discriminant analysis with these two quantities as variables. The rate of correct detection is 80% on average.

This paper presents a novel feature extraction algorithm based on particle swarms for processing hyperspectral imagery data. Particle swarm optimization, originally developed for global optimization over continuous spaces, is extended to deal with the problem of feature extraction. A formulation utilizing two swarms of particles was developed to optimize simultaneously a desired performance criterion and the number of selected features. Candidate feature sets were evaluated on a regression problem. Artificial neural networks were trained to construct linear and nonlinear models of chemical concentration of glucose in soybean crops. Experimental results utilizing real-world hyperspectral datasets demonstrate the viability of the method. The particle swarms-based approach presented superior performance in comparison with conventional feature extraction methods, on both linear and nonlinear models.

We propose a new genetic fuzzy discretization method with feature selection for the pattern classification problems. Traditional discretization methods categorize a continuous attribute into a number of bins. Because they are made on crisp discretization, there exists considerable information loss. Fuzzy discretization allows overlapping intervals and reflects linguistic classification. However, the number of intervals, the boundaries of intervals, and the degrees of overlapping are intractable to get optimized and a discretization process increases the total amount of data being transformed. We use a genetic algorithm with feature selection not only to optimize these parameters but also to reduce the amount of transformed data by filtering the unconcerned attributes. Experimental results showed considerable improvement on the classification accuracy over a crisp discretization and a typical fuzzy discretization with feature selection.

In this paper, a new subspace-based speech enhancement algorithm is presented. First, we construct a perceptual filterbank from psycho-acoustic model and incorporate it in the subspace-based enhancement approach. This filterbank is created through a five-level wavelet packet decomposition. The masking properties of the human auditory system are then derived based on the perceptual filterbank. Finally, the prior SNR and the masking threshold of each critical band are taken to decide the attenuation factor of the optimal linear estimator. Five different types of in-car noises in TAICAR database were used in our evaluation. The experimental results demonstrated that our approach outperformed conventional subspace and spectral subtraction methods.

This paper describes a morpheme-based pronunciation model that is especially useful to develop the pronunciation lexicon for Large Vocabulary Continuous Speech Recognition (LVCSR) in Korean. To address pronunciation variation in Korean, we analyze phonological rules based on phonemic contexts together with morphological category and morpheme boundary information. Since the same phoneme sequences can be pronounced in different ways at across morpheme boundary, incorporating morphological environment is required to manipulate pronunciation variation modeling. We implement a rule-based pronunciation variants generator to produce a pronunciation lexicon with context-dependent multiple variants. At the lexical level, we apply an explicit modeling of pronunciation variation to add pronunciation variants at across morphemes as well as within morpheme into the pronunciation lexicon. At the acoustic level, we train the phone models with re-labeled transcriptions through forced alignment using context-dependent pronunciation lexicon. The proposed pronunciation lexicon offers the potential benefit for both training and decoding of a LVCSR system. Subsequently, we perform the speech recognition experiment on read speech task with 34K-morpheme vocabulary. Experiment confirms that improved performance is achieved by pronunciation variation modeling based on morpho-phonological analysis.

This paper investigates the use of the affine transformation matrix when employing principal component analysis (PCA) to compress the data of 3D animation models. Satisfactory results were achieved for the common 3D models by using PCA because it can simplify several related variables to a few independent main factors, in addition to making the animation identical to the original by using linear combinations. The selection of the principal component factor (also known as the base) is still a subject for further research. Selecting a large number of bases could improve the precision of the animation and reduce distortion for a large data volume. Hence, a formula is required for base selection. This study develops an automatic PCA selection method, which includes the selection of suitable bases and a PCA separately on the three axes to select the number of suitable bases for each axis. PCA is more suitable for animation models for apparent stationary movement. If the original animation model is integrated with transformation movements such as translation, rotation, and scaling (RTS), the resulting animation model will have a greater distortion in the case of the same base vector with regard to apparent stationary movement. This paper is the first to extract the model movement characteristics using the affine transformation matrix and then to compress 3D animation using PCA. The affine transformation matrix can record the changes in the geometric transformation by using 44 matrices. The transformed model can eliminate the influences of geometric transformations with the animation model normalized to a limited space. Subsequently, by using PCA, the most suitable base vector (variance) can be selected more precisely.

A hierarchical representation formed by an octree for a volume ray casting is a well-known data structure to skip over transparent regions requiring little preprocessing and storage. However, it accompanies unnecessary comparison and level shift between octants. We propose a new data structure named half-skewed octree, which is an auxiliary octree to support the conventional octree. In preprocessing step, a half-skewed octree selects eight different child octants in each generation step compared with the conventional octree. During rendering, after comparing an octant of the conventional octree with corresponding octant of the half-skewed octree simultaneously at the same level, a ray chooses one of two octants to jump over transparent regions farther away. By this method, we can reduce unnecessary comparison and level shift between octants. Another problem of a conventional octree structure is that it is difficult to determine a distance from the boundary of a transparent octant to opposite boundary. Although we exploit the previously proposed distance template, we cannot expect the acceleration when a ray direction is almost parallel to the octant's boundary. However, our method can solve it without additional operations because a ray selects one octant to leap farther away. As a result, our approach is much faster than the method using conventional octree while preserving image quality and requiring minimal storage.

In this paper, we propose a learning classifier based on maximum entropy (ME) for resolving zero-anaphora in Chinese text. Besides regular grammatical, lexical, positional and semantic features motivated by previous research on anaphora resolution, we develop two innovative Web-based features for extracting additional semantic information from the Web. The values of the two features can be obtained easily by querying the Web using some patterns. Our study shows that our machine learning approach is able to achieve an accuracy comparable to that of state-of-the-art systems. The Web as a knowledge source can be incorporated effectively into the ME learning framework and significantly improves the performance of our approach.

In this paper, we describe a two-phase method for biomedical named entity recognition consisting of term boundary detection and biomedical category labeling. The term boundary detection can be defined as a task to assign label sequences to a given sentence, and biomedical category labeling can be viewed as a local classification problem which does not need knowledge of the labels of other named entities in a sentence. The advantage of dividing the recognition process into two phases is that we can measure the effectiveness of models at each phase and select separately the appropriate model for each subtask. In order to obtain a better performance in biomedical named entity recognition, we conducted comparative experiments using several learning methods at each phase. Moreover, results by these machine learning based models are refined by rule-based postprocessing. We tested our methods on the JNLPBA 2004 shared task and the GENIA corpus.

Traditional load shedding algorithms for data stream systems calculate current operator selectivity over several run periods and use them to determine where to shed load during the next run period. In this paper, we show that the current selectivity may change due to the implementation of load shedding. Our algorithm, called RLS, determines the optimum drop location by these changed selectivity rather than those pre-calculated values. Simulation results demonstrate that RLS achieves higher accuracy than traditional algorithms.

This letter presents an improved visible watermarking scheme for halftone images. It incorporates watermark embedding into ordered dither halftoning by threshold modulation. The input images include a continuous-tone host image (e.g. an 8-bit gray level image) and a binary watermark image, and the output is a halftone image with a visible watermark. Our method is content adaptive because it takes local intensity information of the host image into account. Experimental results demonstrate effectiveness of the proposed technique. It can be used in practical applications for halftone images, such as commercial advertisement, content annotation, copyright announcement, etc.

This paper presents an inversion algorithm for dynamic Bayesian networks towards robust speech recognition, namely DBNI, which is a generalization of hidden Markov model inversion (HMMI). As a dual procedure of expectation maximization (EM)-based model reestimation, DBNI finds the 'uncontaminated' speech by moving the input noisy speech to the Gaussian means under the maximum likelihood (ML) sense given the DBN models trained on clean speech. This algorithm can provide both the expressive advantage from DBN and the noise-removal feature from model inversion. Experiments on the Aurora 2.0 database show that the hidden feature model (a typical DBN for speech recognition) with the DBNI algorithm achieves superior performance in terms of word error rate reduction.

Grey world algorithm is a well-known color constancy algorithm. It is based on the Grey-World assumption i.e., the average reflectance of surfaces in the world is achromatic. This algorithm is simple and has low computational costs. However, for the images with several colors, the light source color could not be estimated correctly using the Grey World algorithm. In this paper, we propose a Multi-scale Adaptive Grey World algorithm (MAGW). First, multi-scale images are obtained based on wavelet transformation and the illumination color is estimated from different scales images. Then according to the estimated illumination color, the original image is mapped into the image under a canonical illumination with supervision of an adaptive reliability function, which is based on the image entropy. The experimental results show that our algorithm is effective and also has low computational costs.