This paper investigates a new method for fusion of scores generated by multiple classification sub-systems that help to further reduce the classification error rate in Spoken Language Recognition (SLR). In recent studies, a variety of effective classification algorithms have been developed for SLR. Hence, it has been a common practice in the National Institute of Standards and Technology (NIST) Language Recognition Evaluations (LREs) to fuse the results from several classification sub-systems to boost the performance of the SLR systems. In this work, we introduce a discriminative performance measure to optimize the performance of the fusion of 7 language classifiers developed as IIR's submission to the 2009 NIST LRE. We present an Error Corrective Fusion (ECF) method in which we iteratively learn the fusion weights to minimize error rate of the fusion system. Experiments conducted on the 2009 NIST LRE corpus demonstrate a significant improvement compared to individual sub-systems. Comparison study is also conducted to show the effectiveness of the ECF method.

Petri net is a powerful modeling tool for concurrent systems. Subclasses of Petri net are suggested to model certain realistic applications with less computational cost. Structurally weakly persistent net (SWPN) is one of such subclasses where liveness is verified in deterministic polynomial time. This paper studies the computational complexity to verify whether a give net is SWPN. 3UNSAT problem is reduced to the problem to verify whether a net is not SWPN. This implies co-NP completeness of verification problem of SWPN.

Object tracking is a major technique in image processing and computer vision. Tracking speed will directly determine the quality of applications. This paper presents a parallel implementation for a recently proposed scale- and rotation-invariant on-line object tracking system. The algorithm is based on NVIDIA's Graphics Processing Units (GPU) using Compute Unified Device Architecture (CUDA), following the model of single instruction multiple threads. Specifically, we analyze the original algorithm and propose the GPU-based parallel design. Emphasis is placed on exploiting the data parallelism and memory usage. In addition, we apply optimization technique to maximize the utilization of NVIDIA's GPU and reduce the data transfer time. Experimental results show that our GPGPU-based method running on a GTX480 graphics card could achieve up to 12X speed-up compared with the efficiency equivalence on an Intel E8400 3.0 GHz CPU, including I/O time.

Measuring the relational similarity between word pairs is important in numerous natural language processing tasks such as solving word analogy questions, classifying noun-modifier relations and disambiguating word senses. We propose a supervised classification method to measure the similarity between semantic relations that exist between words in two word pairs. First, each pair of words is represented by a vector of automatically extracted lexical patterns. Then a binary Support Vector Machine is trained to recognize word pairs with similar semantic relations to a given word pair. To train and evaluate the proposed method, we use a benchmark dataset that contains 374 SAT multiple-choice word-analogy questions. To represent the relations that exist between two word pairs, we experiment with 11 different feature functions, including both symmetric and asymmetric feature functions. Our experimental results show that the proposed method outperforms several previously proposed relational similarity measures on this benchmark dataset, achieving an SAT score of 46.9.

Previous studies of pattern recognition have shown that classifiers ensemble approaches can lead to better recognition results. In this paper, we apply the voting technique for the CoNLL-2010 shared task on detecting hedge cues and their scope in biomedical texts. Six machine learning-based systems are combined through three different voting schemes. We demonstrate the effectiveness of classifiers ensemble approaches and compare the performance of three different voting schemes for hedge cue and their scope detection. Experiments on the CoNLL-2010 evaluation data show that our best system achieves an F-score of 87.49% on hedge detection task and 60.87% on scope finding task respectively, which are significantly better than those of the previous systems.

We propose multiscale bagging as a modification of the bagging procedure. In ordinary bagging, the bootstrap resampling is used for generating bootstrap samples. We replace it with the multiscale bootstrap algorithm. In multiscale bagging, the sample size m of bootstrap samples may be altered from the sample size n of learning dataset. For assessing the output of a classifier, we compute bootstrap probability of class label; the frequency of observing a specified class label in the outputs of classifiers learned from bootstrap samples. A scaling-law of bootstrap probability with respect to σ2=n/m has been developed in connection with the geometrical theory. We consider two different ways for using multiscale bagging of classifiers. The first usage is to construct a confidence set of class labels, instead of a single label. The second usage is to find inputs close to decision boundaries in the context of query by bagging for active learning. It turned out, interestingly, that an appropriate choice of m is m =-n, i.e., σ2=-1, for the first usage, and m =∞, i.e., σ2=0, for the second usage.

This paper addresses the problem of ordering the color SIFT descriptors in the independent component analysis for image classification. Component ordering is of great importance for image classification, since it is the foundation of feature selection. To select distinctive and compact independent components (IC) of the color SIFT descriptors, we propose two ordering approaches based on local variation, named as the localization-based IC ordering and the sparseness-based IC ordering. We evaluate the performance of proposed methods, the conventional IC selection method (global variation based components selection) and original color SIFT descriptors on object and scene databases, and obtain the following two main results. First, the proposed methods are able to obtain acceptable classification results in comparison with original color SIFT descriptors. Second, the highest classification rate can be obtained by using the global selection method in the scene database, while the local ordering methods give the best performance for the object database.

Boosting over weak classifiers is widely used in pedestrian detection. As the number of weak classifiers is large, researchers always use a sampling method over weak classifiers before training. The sampling makes the boosting process harder to reach the fixed target. In this paper, we propose a partial derivative guidance for weak classifier mining method which can be used in conjunction with a boosting algorithm. Using weak classifier mining method makes the sampling less degraded in the performance. It has the same effect as testing more weak classifiers while using acceptable time. Experiments demonstrate that our algorithm can process quicker than [1] algorithm in both training and testing, without any performance decrease. The proposed algorithms is easily extending to any other boosting algorithms using a window-scanning style and HOG-like features.

A novel approach for detecting anomaly in visual surveillance system is proposed in this paper. It is composed of three parts: (a) a dense motion field and motion statistics method, (b) motion directional PCA for feature dimensionality reduction, (c) an improved one-class SVM for one-class classification. Experiments demonstrate the effectiveness of the proposed algorithm in detecting abnormal events in surveillance video, while keeping a low false alarm rate. Our scheme works well in complicated situations that common tracking or detection modules cannot handle.

A 10-Gbit/s-class ac-coupled average-detection-type burst-mode receiver (B-Rx) with an ultra fast response and a high tolerance to the long consecutive identical digits has been developed. Key features of the circuit design are the baseline-wander common-mode rejection technique and the inverted distortion technique adopted in the limiting amplifier to cope with both the fast response and the high tolerance. Our B-Rx with newly developed limiting amplifier IC achieved a settling time of less than 150 ns, a sensitivity of -29.8 dBm, and a dynamic range of 23.8 dB with a 231-1 pseudo random bit sequences. Moreover, we also describe several potential B-Rx applications. We achieved better performance by applying the proposed systems to our B-Rx.

A novel predistortion technique using an automatic average-power bias controlled diode is proposed to compensate the complicated nonlinear characteristics of a microwave class-F power amplifier using an AlGaN/GaN HEMT. The optimum value for diode bias voltage is automatically set according to detected input average RF power level. A high-efficiency 1.9 GHz class-F GaN HEMT power amplifier with the automatic average-power bias control (ABC) diode linearizer achieves an improved third order inter-modulation distortion (IMD3) of better than -45 dBc at a smaller than 6 dB output power back-off from a saturated output power of 27 dBm, without changing drain efficiency. The adjacent channel leakage power ratio (ACPR) for 1.9 GHz W-CDMA signals is below -40 dBc at output power levels of smaller than 20 dBm for the class-F power amplifier.

This letter presents a new automatic musical genre classification method based on an informative song-level representation, in which the mutual information between the feature and the genre label is maximized. By efficiently combining distance-based indexing with informative features, the proposed method represents a song as one vector instead of complex statistical models. Experiments on an audio genre DB show that the proposed method can achieve the classification accuracy comparable or superior to the state-of-the-art results.

This paper presents an efficient yet powerful codebook model, named classified codebook model, to categorize natural scene category. The current codebook model typically resorts to large codebook to obtain higher performance for scene categorization, which severely limits the practical applicability of the model. Our model formulates the codebook model with the theory of vector quantization, and thus uses the famous technique of classified vector quantization for scene-category modeling. The significant feature in our model is that it is beneficial for scene categorization, especially at small codebook size, while saving much computation complexity for quantization. We evaluate the proposed model on a well-known challenging scene dataset: 15 Natural Scenes. The experiments have demonstrated that our model can decrease the computation time for codebook generation. What is more, our model can get better performance for scene categorization, and the gain of performance becomes more pronounced at small codebook size.

The least-squares probabilistic classifier (LSPC) is a computationally-efficient alternative to kernel logistic regression. However, to assure its learned probabilities to be non-negative, LSPC involves a post-processing step of rounding up negative parameters to zero, which can unexpectedly influence classification performance. In order to mitigate this problem, we propose a simple alternative scheme that directly rounds up the classifier's negative outputs, not negative parameters. Through extensive experiments including real-world image classification and audio tagging tasks, we demonstrate that the proposed modification significantly improves classification accuracy, while the computational advantage of the original LSPC remains unchanged.

In this letter, a new terrain type classifier is proposed for polarimetric Synthetic Aperture Radar (Pol-SAR) images. This classifier uses the binary tree structure. The homogenous and inhomogeneous areas are first classified by the support vector machine (SVM) classifier based on the texture features extracted from the span image. Then the homogenous and inhomogeneous areas are, respectively, classified by the traditional Wishart classifier and the SVM classifier based on the texture features. Using a NASA/JPL AIRSAR image, the authors achieve the classification accuracy of up to 98%, demonstrating the effectiveness of the proposed method.

This paper presents various types of iterative progressive numerical methods (IPNMs) for the computation of electromagnetic (EM) wave scattering from many objects and reports comparatively the performance of these methods. The original IPNM is similar to the Jacobi method which is one of the classical linear iterative solvers. Then the modified IPNMs are based on other classical solvers like the Gauss-Seidel (GS), the relaxed Jacobi, the successive overrelaxation (SOR), and the symmetric SOR (SSOR) methods. In the original and modified IPNMs, we repeatedly solve linear systems of equations by using a nonstationary iterative solver. An initial guess and a stopping criterion are discussed in order to realize a fast computation. We treat EM wave scattering from 27 perfectly electric conducting (PEC) spheres and evaluate the performance of the IPNMs. However, the SOR- and SSOR-type IPNMs are not subject to the above numerical test in this paper because an optimal relaxation parameter is not possible to determine in advance. The evaluation reveals that the IPNMs converge much faster than a standard BEM computation. The relaxed Jacobi-type IPNM is better than the other types in terms of the net computation time and the application range for the distance between objects.

This paper presents an improved Gini-Index algorithm to correct feature-selection bias in text classification. Gini-Index has been used as a split measure for choosing the most appropriate splitting attribute in decision tree. Recently, an improved Gini-Index algorithm for feature selection, designed for text categorization and based on Gini-Index theory, was introduced, and it has proved to be better than the other methods. However, we found that the Gini-Index still shows a feature selection bias in text classification, specifically for unbalanced datasets having a huge number of features. The feature selection bias of the Gini-Index in feature selection is shown in three ways: 1) the Gini values of low-frequency features are low (on purity measure) overall, irrespective of the distribution of features among classes, 2) for high-frequency features, the Gini values are always relatively high and 3) for specific features belonging to large classes, the Gini values are relatively lower than those belonging to small classes. Therefore, to correct that bias and improve feature selection in text classification using Gini-Index, we propose an improved Gini-Index (I-GI) algorithm with three reformulated Gini-Index expressions. In the present study, we used global dimensionality reduction (DR) and local DR to measure the goodness of features in feature selections. In experimental results for the I-GI algorithm, we obtained unbiased feature values and eliminated many irrelevant general features while retaining many specific features. Furthermore, we could improve the overall classification performances when we used the local DR method. The total averages of the classification performance were increased by 19.4 %, 15.9 %, 3.3 %, 2.8 % and 2.9 % (kNN) in Micro-F1, 14 %, 9.8 %, 9.2 %, 3.5 % and 4.3 % (SVM) in Micro-F1, 20 %, 16.9 %, 2.8 %, 3.6 % and 3.1 % (kNN) in Macro-F1, 16.3 %, 14 %, 7.1 %, 4.4 %, 6.3 % (SVM) in Macro-F1, compared with tf*idf, χ2, Information Gain, Odds Ratio and the existing Gini-Index methods according to each classifier.

In this letter, we give a generalized construction for sets of frequency-hopping sequences (FHSs) based on power-residue sequences. Our construction encompasses a known optimal construction and can generate new optimal sets of FHSs which simultaneously achieve the Peng-Fan bound and the Lempel-Greenberger bound.

This study shows a fast simulation method for turbo codes over an additive white class A noise (AWAN) channel. The reduction of the estimation time is achieved by applying importance sampling (IS) which is one of the variance reduction simulation methods. In order to adapt the AWAN channel, we propose a design method of a simulation probability density function (PDF) utilized in IS. The proposed simulation PDF is related to the Bhattacharyya bound to evaluate wider area of the signal space than the conventional method. Since the mean translation method, which is a conventional design method of the simulation PDF used in IS, is optimized for an additive white Gaussian noise channel, the evaluation time of the error performance of turbo codes over the AWAN channel can not be reduced. To evaluate BER of 10-8, the simulation time of the proposed method can be reduced to 1/104 under the condition of the same accuracy of the traditional Monte Carlo simulation method.

This paper presents an unsupervised learning-based method for selection of feature points and object category classification without previous setting of the number of categories. Our method consists of the following procedures: 1)detection of feature points and description of features using a Scale-Invariant Feature Transform (SIFT), 2)selection of target feature points using One Class-Support Vector Machines (OC-SVMs), 3)generation of visual words of all SIFT descriptors and histograms in each image of selected feature points using Self-Organizing Maps (SOMs), 4)formation of labels using Adaptive Resonance Theory-2 (ART-2), and 5)creation and classification of categories on a category map of Counter Propagation Networks (CPNs) for visualizing spatial relations between categories. Classification results of static images using a Caltech-256 object category dataset and dynamic images using time-series images obtained using a robot according to movements respectively demonstrate that our method can visualize spatial relations of categories while maintaining time-series characteristics. Moreover, we emphasize the effectiveness of our method for category classification of appearance changes of objects.