Search

Different from the common approaches that use either hypothesis test or classifier for biomarker discovery, we applied the integrative hypothesis test (IHT) that combined both to identifying miRNAs for differentiation between lung cancer and Chronic Obstructive Pulmonary Disease (shortly L-C differentiation) on GEO data set GSE24709, and further extended IHT implementation by bootstrapping aided ranking and mean-variance based reliability check, which outputs a list of the top-15 differentially expressed miRNAs that confirmed the previously reported 14 miRNAs for L-C differentiation from a very different perspective plus an additional one. Moreover, we conducted a literature survey for a further explanation via dividing the 15 miRNAs into subclasses based on known relevances to the two diseases. Also, every pair of 15 miRNAs is exhaustively examined on their joint effect via p -value, misclassification, and correlation, which identifies core pairs and linked cliques as joint miRNAs biomarkers. [ABSTRACT FROM AUTHOR]

In order to seek non-propagation method to train generalized single-hidden layer feed forward neural networks, extreme learning machine was proposed, which has been proven to be an effective and efficient model for both multi-class classification and regression. Different from most of existing studies which consider extreme learning machine as a classifier, we make improvements on it to let it become a feature extraction model in this paper. Specifically, a discriminative extreme learning machine with supervised sparsity preserving (SPELM) model is proposed. From the hidden layer to output layer, SPELM performs as a subspace learning method by considering the discriminative as well as sparsity information of data. The sparsity information of data is identified by solving a supervised sparse representation objective. Experiments are conducted on four widely used image benchmark data sets and the classification results demonstrate the effectiveness of the proposed SPELM model. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) uses a non-iterative method to train single-hidden-layer feed-forward networks (SLFNs), which has been proven to be an efficient and effective learning model for both classification and regression. The main advantage of ELM lies in that the input weights as well as the hidden layer biases can be randomly generated, which contributes to the analytical solution of output weights. In this paper, we propose a discriminative manifold ELM (DMELM) by simultaneously considering the discriminative information and geometric structure of data; specifically, we exploit the discriminative information in the local neighborhood around each data point. To this end, a graph regularizer based on a newly designed graph Laplacian to characterize both properties is formulated and incorporated into the ELM objective. In DMELM, the output weights can also be obtained in analytical form. Extensive experiments are conducted on image and EEG signal classification to evaluate the effectiveness of DMELM. The results show that DMELM consistently achieves better performance than original ELM and yields promising results in comparison with several state-of-the-art algorithms, which suggests that both the discriminative as well as manifold information are beneficial to classification. [ABSTRACT FROM AUTHOR]

Constructing an informative and discriminative graph plays an important role in various pattern recognition tasks such as clustering and classification. Among the existing graph-based learning models, low-rank representation (LRR) is a very competitive one, which has been extensively employed in spectral clustering and semi-supervised learning (SSL). In SSL, the graph is composed of both labeled and unlabeled samples, where the edge weights are calculated based on the LRR coefficients. However, most of existing LRR related approaches fail to consider the geometrical structure of data, which has been shown beneficial for discriminative tasks. In this paper, we propose an enhanced LRR via sparse manifold adaption, termed manifold low-rank representation (MLRR), to learn low-rank data representation. MLRR can explicitly take the data local manifold structure into consideration, which can be identified by the geometric sparsity idea; specifically, the local tangent space of each data point was sought by solving a sparse representation objective. Therefore, the graph to depict the relationship of data points can be built once the manifold information is obtained. We incorporate a regularizer into LRR to make the learned coefficients preserve the geometric constraints revealed in the data space. As a result, MLRR combines both the global information emphasized by low-rank property and the local information emphasized by the identified manifold structure. Extensive experimental results on semi-supervised classification tasks demonstrate that MLRR is an excellent method in comparison with several state-of-the-art graph construction approaches. [ABSTRACT FROM AUTHOR]

Abstract: For many human machine interaction systems, techniques for continuously estimating the vigilance of operators are highly desirable to ensure work safety. Up to now, various signals are studied for vigilance analysis. Among them, electroencephalogram (EEG) is the most commonly used signal. In this paper, extreme learning machine (ELM) and its modifications with L 1 norm and L 2 norm penalties are adopted for EEG-based vigilance estimation. A comparative study on system performance is conducted among ordinary ELM, its modifications, and support vector machines (SVMs). Experimental results show that, compared with SVMs, the ordinary ELM and its modifications can all dramatically speed up the training process while still achieving similar or better vigilance estimation accuracy. In addition, the following three observations have been made from the experiment results: (a) the ordinary ELM and the ELM with L 1 norm penalty (LARS-ELM) are sensitive on the number of hidden nodes; (b) the ELM with L 2 norm penalty (regularized-ELM) and the ELMs with both L 1 norm and L 2 norm penalties (LARS-EN-ELM, TROP-ELM) are stable and insensitive on the number of hidden nodes; and (c) regularized-ELM has a much faster training speed, while LARS-EN-ELM can achieve better vigilance estimation accuracy. [Copyright &y& Elsevier]

Abstract: In this paper, we propose a support vector machine with automatic confidence (SVMAC) for pattern classification. The main contributions of this work to learning machines are twofold. One is that we develop an algorithm for calculating the label confidence value of each training sample. Thus, the label confidence values of all of the training samples can be considered in training support vector machines. The other one is that we propose a method for incorporating the label confidence value of each training sample into learning and derive the corresponding quadratic programming problems. To demonstrate the effectiveness of the proposed SVMACs, a series of experiments are performed on three benchmarking pattern classification problems and a challenging gender classification problem. Experimental results show that the generalization performance of our SVMACs is superior to that of traditional SVMs. [Copyright &y& Elsevier]

This article explores the interdependences between subcellular locations and incorporates them with support vector machines for prediction of protein subcellular localisation. Traditional prediction systems utilise a 'flat' structure of classifiers, such as the one-versus-all and one-versus-one schemes, with amino acid compositions to perform the prediction. Apart from those existing studies that ignore the interdependences between subcellular locations, we take advantage of a hierarchical structure to organise the subcellular locations and model their relationships. Here, we propose to use four kinds of hierarchical prediction methods and make comparative studies on three datasets. Experimental results show that three of the hierarchical models outperform the traditional 'flat' model in terms of tree loss values. In particular, one hierarchical model outperforms the traditional 'flat' model for all evaluation measures. Moreover, we gained some valuable insights into the sorting process by using hierarchical structures. [ABSTRACT FROM AUTHOR]

Prediction of protein subcellular localization is an important issue in computational biology because it provides important clues for the characterization of protein functions. Currently, much research has been dedicated to developing automatic prediction tools. Most, however, focus on mono-locational proteins, i.e., they assume that proteins exist in only one location. It should be noted that many proteins bear multi-locational characteristics and carry out crucial functions in biological processes. This work aims to develop a general pattern classifier for predicting multiple subcellular locations of proteins. We use an ensemble classifier, called the min-max modular support vector machine (M3-SVM), to solve protein subcellular multi-localization problems; and, propose a module decomposition method based on gene ontology (GO) semantic information for M3-SVM. The amino acid composition with secondary structure and solvent accessibility information is adopted to represent features of protein sequences. We apply our method to two multi-locational protein data sets. The M3-SVMs show higher accuracy and efficiency than traditional SVMs using the same feature vectors. And the GO decomposition also helps to improve prediction accuracy. Moreover, our method has a much higher rate of accuracy than existing subcellular localization predictors in predicting protein multi-localization. [ABSTRACT FROM AUTHOR]

Principal components analysis (PCA) is a popular linear feature extractor, and widely used in signal processing, face recognition, etc. However, axes of the lower-dimensional space, i.e., principal components, are a set of new variables carrying no clear physical meanings. Thus we propose unsupervised feature selection algorithms based on eigenvectors analysis to identify critical original features for principal component. The presented algorithms are based on k-nearest neighbor rule to find the predominant row components and eight new measures are proposed to compute the correlation between row components in transformation matrix. Experiments are conducted on benchmark data sets and facial image data sets for gender classification to show their superiorities. [ABSTRACT FROM AUTHOR]

Abstract: The problem of selecting a subset of relevant features is classic and found in many branches of science including—examples in pattern recognition. In this paper, we propose a new feature selection criterion based on low-loss nearest neighbor classification and a novel feature selection algorithm that optimizes the margin of nearest neighbor classification through minimizing its loss function. At the same time, theoretical analysis based on energy-based model is presented, and some experiments are also conducted on several benchmark real-world data sets and facial data sets for gender classification to show that the proposed feature selection method outperforms other classic ones. [Copyright &y& Elsevier]

Abstract: The image Euclidean distance (IMED) considers the spatial relationship between the pixels of different images and can easily be embedded in existing image recognition algorithms that are based on Euclidean distance. IMED uses the prior knowledge that pixels located near one another have little variance in gray scale values, and defines a metric matrix according to the spatial distance between pixels. In this paper, we propose an adaptive image Euclidean distance (AIMED), which considers not only the prior spatial knowledge, but also the prior gray level knowledge from images. The most important advantage of the proposed AIMED over IMED is that AIMED makes the metric matrix adaptive to the content of the concerned images. Two ways of using gray level information are proposed. One is based on gray level distances, and the other is based on cosine dissimilarity of gray levels. Experiments on two facial databases and a handwritten digital database show that AIMED achieves the highest classification accuracy when it is embedded in nearest neighbor classifiers, principal component analysis, and support vector machines. [Copyright &y& Elsevier]

In this paper, we present a novel method for multi-view gender classification considering both shape and texture information to represent facial images. The face area is divided into small regions from which local binary pattern (LBP) histograms are extracted and concatenated into a single vector efficiently representing a facial image. Following the idea of local binary pattern, we propose a new feature extraction approach called multi-resolution LBP, which can retain both fine and coarse local micro-patterns and spatial information of facial images. The classification tasks in this work are performed by support vector machines (SVMs). The experiments clearly show the superiority of the proposed method over both support gray faces and support Gabor faces on the CAS-PEAL face database. A higher correct classification rate of 96.56% and a higher cross validation average accuracy of 95.78% have been obtained. In addition, the simplicity of the proposed method leads to very fast feature extraction, and the regional histograms and fine-to-coarse description of facial images allow for multi-view gender classification. [ABSTRACT FROM AUTHOR]

In this paper we present a method for converting general nonlinear programming (NLP) problems into separable programming (SP) problems by using feedforward neural networks (FNNs). The basic idea behind the method is to use two useful features of FNNs: their ability to approximate arbitrary continuous nonlinear functions with a desired degree of accuracy and their ability to express nonlinear functions in terms of parameterized compositions of functions of single variables. According to these two features, any nonseparable objective functions and/or constraints in NLP problems can be approximately expressed as separable functions with FNNs. Therefore, any NLP problems can be converted into SP problems. The proposed method has three prominent features. (a) It is more general than existing transformation techniques; (b) it can be used to formulate optimization problems as SP problems even when their precise analytic objective function and/or constraints are unknown; (c) the SP problems obtained by the proposed method may highly facilitate the selection of grid points for piecewise linear approximation of nonlinear functions. We analyze the computational complexity of the proposed method and compare it with an existing transformation approach. We also present several examples to demonstrate the method and the performance of the simplex method with the restricted basis entry rule for solving SP problems. [Copyright &y& Elsevier]

Provides information on a study which illustrated a method for dealing with the inverse problem by using a mathematical programming techniques. Statement of the inverse problem; Computing inversions by nonlinear programming techniques; Computing inversions by separable programming; Relations between inversions and network parameters.

Presents information on a study which proposed a method for decomposing pattern classification problems based on the class relations among training data. Difficulties in using artificial neural networks; Issues to address to implement modular learning; Methodology; Related literature; Conclusions.

Extreme Learning Machine (ELM), which was initially proposed for training single-layer feed-forward networks (SLFNs), provides us a unified efficient and effective framework for regression and multiclass classification. Though various ELM variants were proposed in recent years, most of them focused on the supervised learning scenario while little effort was made to extend it into unsupervised learning paradigm. Therefore, it is of great significance to put ELM into learning tasks with only unlabeled data. One popular approach for mining knowledge from unlabeled data is based on the manifold assumption, which exploits the geometrical structure of data by assuming that nearby points will also be close to each other in transformation space. However, considering the manifold information only is insufficient for discriminative tasks. In this paper, we propose an improved unsupervised discriminative ELM (UDELM) model, whose main advantage is to combine the local manifold learning with global discriminative learning together. UDELM can be efficiently optimized by solving a generalized eigen-value decomposition problem. Extensive comparisons over several state-of-the-art models on clustering image and emotional EEG data demonstrate the efficacy of UDELM. [ABSTRACT FROM AUTHOR]

Abstract: Vigilance level estimation can be used to prevent disastrous accident occurring frequently in high-risk tasks. Electroencephalograph (EEG) based Brain Computer Interface (BCI) is one of the most important tools for detecting one's brain electrical activities. Unfortunately, several problems including its sensitivity to artifacts, inaccurate labels and the great diversity of patterns within EEG signals present great challenges to predict vigilance level reliably. In this paper we propose an integrated approach to estimate vigilance level, which incorporates an automatically artifact removing preprocess, a novel vigilance labeling method and finally a Gaussian Mixed Model (GMM) to discover the underlying pattern of EEG signals. Extensive off-line experiments are conducted on 12 groups of data sets to show the effectiveness of our integrated approach in the real-time application. A reasonably high classification performance (88.46% over 12 data sets) is obtained with low delay by employing only one channel in the frontal lobe, which is in accordance with the conclusions of brain science and is of significance in practice. [Copyright &y& Elsevier]

Abstract: Recently, emotion classification from EEG data has attracted much attention with the rapid development of dry electrode techniques, machine learning algorithms, and various real-world applications of brain–computer interface for normal people. Until now, however, researchers had little understanding of the details of relationship between different emotional states and various EEG features. To improve the accuracy of EEG-based emotion classification and visualize the changes of emotional states with time, this paper systematically compares three kinds of existing EEG features for emotion classification, introduces an efficient feature smoothing method for removing the noise unrelated to emotion task, and proposes a simple approach to tracking the trajectory of emotion changes with manifold learning. To examine the effectiveness of these methods introduced in this paper, we design a movie induction experiment that spontaneously leads subjects to real emotional states and collect an EEG data set of six subjects. From experimental results on our EEG data set, we found that (a) power spectrum feature is superior to other two kinds of features; (b) a linear dynamic system based feature smoothing method can significantly improve emotion classification accuracy; and (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. [Copyright &y& Elsevier]

Recent large-scale hierarchical classification tasks typically have tens of thousands of classes on which the most widely used approach to multiclass classification--one-versus-rest--becomes intractable due to computational complexity. The top-down methods are usually adopted instead, but they are less accurate because of the so-called error-propagation problem in their classifying phase. To address this problem, this paper proposes a meta-top-down method that employs metaclassification to enhance the normal top-down classifying procedure. The proposed method is first analyzed theoretically on complexity and accuracy, and then applied to five real-world large-scale data sets. The experimental results indicate that the classification accuracy is largely improved, while the increased time costs are smaller than most of the existing approaches. [ABSTRACT FROM AUTHOR]

In this paper, we propose a multi-view gender classification system with a hierarchical framework using facial images as input. The front end of the framework is a classifier, which will properly divides the input images into several groups. To ease the data sparsity problem in the multi-view scenario, facial symmetry is used to reduce the number of views. Moreover, we adopt soft assignment when dividing the input data, which can reduce the errors caused by the boundary effect in hard assignment. Then for each group, we train a gender classifier, called an expert. These experts can be any commonly used classifiers, such as support vector machines or neural networks. In this step, facial components instead of the whole face are used to achieve higher robustness against variations caused by facial alignment, illumination and occlusions. Experimental results demonstrate that our framework significantly improves the performance. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we propose a novel gender classification framework, which utilizes not only facial features, but also external information, i.e. hair and clothing. Instead of using the whole face, we consider five facial components: forehead, eyes, nose, mouth and chin. We also design feature extraction methods for hair and clothing; these features have seldom been used in previous work because of their large variability. For each type of feature, we train a single support vector machine classifier with probabilistic output. The outputs of these classifiers are combined using various strategies, namely fuzzy integral, maximal, sum, voting, and product rule. The major contributions of this paper are (1) investigating the gender discriminative ability of clothing information; (2) using facial components instead of the whole face to obtain higher robustness for occlusions and noise; (3) exploiting hair and clothing information to facilitate gender classification. Experimental results show that our proposed framework improves classification accuracy, even when images contain occlusions, noise, and illumination changes. [Copyright &y& Elsevier]

In electroencephalography (EEG)-based affective brain–computer interfaces (aBCIs), there is a consensus that EEG features extracted from different frequency bands and channels have different abilities in emotion expression. Besides, EEG is so weak and non-stationary that easily causes distribution discrepancies for EEG data collected at different times; therefore, it is necessary to explore the affective activation patterns in cross-session emotion recognition. To address these two problems, we propose a self-weighted semi-supervised classification (SWSC) model in this article for joint EEG-based cross-session emotion recognition and affective activation patterns mining, whose merits include: 1) using both the labeled and unlabeled samples from different sessions for better capturing data characteristics; 2) introducing a self-weighted variable to learn the importance of EEG features adaptively and quantitatively; and 3) mining the activation patterns including the critical EEG frequency bands and channels automatically based on the learned self-weighted variable. Extensive experiments are conducted on the benchmark SEED_IV emotional dataset and SWSC obtained excellent average accuracies of 77.40%, 79.55%, and 81.52% in three cross-session emotion recognition tasks. Moreover, SWSC identifies that the Gamma frequency band contributes the most and the EEG channels in prefrontal, left/right temporal, and (central) parietal lobes are more important for cross-session emotion recognition. [ABSTRACT FROM AUTHOR]

In recent years, sleepiness during driving has become a main cause for traffic accidents. However, the fact is that we know very little yet about the electrophysiological marker for assessing diver sleepiness. Previous studies and our researches have shown that alpha blocking phenomenon and alpha wave attenuation-disappearance phenomenon represent two different sleepiness levels, the relaxed wakefulness and the sleep onset, respectively. This paper proposes a novel model for driver sleepiness detection based on electroencephalography (EEG) and electrooculography (EOG) signals. Our model aims to track the change in alpha waves and differentiate the two alpha-related phenomena. Continuous wavelet transform is adopted to extract features from physiological signals in both time and frequency domains. Meanwhile, Long-Short Term Memory (LSTM) network is introduced to deal with temporal information of EEG and EOG signals. To deal with insufficient physiological sample problem, generative adversarial network (GAN) is used to augment the training dataset. Experimental results indicate that the F1 score for detecting start and end points of alpha waves reaches to around 95%. And Conditional Wasserstein GAN (CWGAN) we adopted was effective in augmenting dataset and boost classifier performance. Meanwhile, our LSTM classifier achieved a mean accuracy of 98% for classifying end points of alpha waves under leave-one-subject-out cross validation. [ABSTRACT FROM AUTHOR]

Extreme Learning Machine (ELM) has been proposed as a new algorithm for training single hidden layer feed forward neural networks. The main merit of ELM lies in the fact that the input weights as well as hidden layer bias are randomly generated and thus the output weights can be obtained analytically, which can overcome the drawbacks incurred by gradient-based training algorithms such as local optima, improper learning rate and low learning speed. Based on the consistency property of data, which enforces similar samples to share similar properties, we propose a discriminative graph regularized Extreme Learning Machine (GELM) for further enhancing its classification performance in this paper. In the proposed GELM model, the label information of training samples are used to construct an adjacent graph and correspondingly the graph regularization term is formulated to constrain the output weights to learn similar outputs for samples from the same class. The proposed GELM model also has a closed form solution as the standard ELM and thus the output weights can be obtained efficiently. Experiments on several widely used face databases show that our proposed GELM can achieve much performance gain over standard ELM and regularized ELM. Moreover, GELM also performs well when compared with the state-of-the-art classification methods for face recognition. [ABSTRACT FROM AUTHOR]

Abstract: Extreme Learning Machine (ELM) as an emergent technology has shown its promising performance in many applications. This paper proposes a parallelized ELM ensemble based on the Min–Max Modular network (M3-network) to meet the challenge of the so-called big data. The proposed M3-ELM first decomposes classification problems into smaller subproblems, then trains an ELM for each subproblem, and in the end ensembles these ELMs with the M3-network. Twelve data sets including both benchmarks and real-world applications are employed to test the proposed method. The experimental results show that M3-ELM not only speeds up the training phrases by 1.6–4.6 times but also reduces the test errors by 0.37–19.51% compared with the normal ELM. The results also indicate that M3-ELM possesses scalability on large-scale tasks and accuracy improvement on imbalanced tasks. [Copyright &y& Elsevier]

Patent classification is a large scale hierarchical text classification (LSHTC) task. Though comprehensive comparisons, either learning algorithms or feature selection strategies, have been fully made in the text categorization field, few work was done for a LSHTC task due to high computational cost and complicated structural label characteristics. For the first time, this paper compares two popular learning frameworks, namely, hierarchical support vector machine (SVM) and k-nearest neighbor ( k-NN) that are applied to a LSHTC task. Our experimental results show that the latter outperforms the former for the LSHTC task, which is quite different from the existing results for normal text categorization tasks. In addition, this paper compares different similarity measures and ranking strategies in k-NN framework for LSHTC task. From our empirical study, conclusions can be drawn that k-NN is more appropriate for the LSHTC task than hierarchical SVM. BM25 outperforms other similarity measures and ListWeak gains a better performance than other ranking strategies. Our empirical results also indicate that using all the labels of the retrieved neighbors can remarkably improve classification performance over only using the first label of the retrieved neighbors. [ABSTRACT FROM AUTHOR]

Abstract: Mn-doped ZnO single crystals have been grown by the flux-Bridgman method from a high temperature solution of 24% ZnO-76% PbF2 (molar fraction). A fast nucleation was induced by the gas-cooling system at the bottom of the Pt crucible. A red Mn:ZnO crystal up to 13 mm×15 mm×(3–4) mm was obtained. Powder XRD patterns showed that the as-grown crystals were in wurtzite structure. Electron Probe Microanalyser (EPMA) revealed that some PbF2 inclusions formed due to fast nucleation. The additional Raman mode at 524 cm?1 was attributed to the incorporation of Mn ions into ZnO crystal. Six-line hyperfine structure was observed in the electron paramagnetic resonance (EPR) spectra, which confirmed that Mn ions occupied Zn2+ lattice sites with valence state of +2. [Copyright &y& Elsevier]