General Classification Problems Research Articles

Objective evaluation of fabric smoothness appearance with an ordinal classification framework based on label noise estimation

Objective fabric smoothness appearance evaluation plays an important role in the textile and apparel industry. In most previous studies, objective fabric smoothness appearance evaluation is defined as a general pattern classification problem. However, the labels in this problem exhibit a natural ordering. Nominal classification ignores the ordinal information, which may cause overfitting in model training. In addition, for the existence of subjective errors, measurement errors, manual errors, etc., the labels in the data might be noisy, which has been rarely discussed previously. This paper proposes an ordinal classification framework based on label noise estimation (OCF-LNE) to objectively evaluate the fabric smoothness appearance degree, which takes the ordinal information and noise of the label in the training data into consideration. The OCF-LNE uses the basic classifier in pre-training as a label noise estimator, and uses the estimated label noise to adjust the labels in further training. The adjusted labels can introduce the ordinal constrain implicitly and reduce the negative impact of label noise in model training. Within a 10 × 10 nested cross-validation, the proposed OCF-LNE achieves 82.86%, 94.29%, and 100% average accuracies under errors of 0, 0.5, and 1 degree, respectively. Experiments on different fabric image features and basic classification models verify the effectiveness of the OCF-LNE. In addition, the proposed method outperforms the state-of-the-art methods for fabric smoothness evaluation and ordinal classification. Promisingly, the OCF-LNE can provide novel ideas for image-based fabric smoothness evaluation.

General multi-view semi-supervised least squares support vector machines with multi-manifold regularization

Multi-view semi-supervised learning achieves great success in recent years which leverages information among labeled and unlabeled multi-view data to improve the generalization performance. So far two classical two-view semi-supervised learning methods are multi-view Laplacian support vector machines (MvLapSVM) and multi-view Laplacian twin support vector machines (MvLapTSVM). But they can only deal with two-view classification problems and cannot deal with general multi-view classification problems. They both solve the quadratic programming problems (QPPs) so that their time complexity is quite high. In this paper, we formulate general multi-view Laplacian least squares support vector machines (GMvLapSVM) and general multi-view Laplacian least squares twin support vector machines (GMvLapTSVM) which solve linear equations as compared to QPPs in MvLapSVM and MvLapTSVM. They can handle the general multi-view classification problems by combining multiple different views in a non-pairwise way. The disagreement among different views is considered as a regularization term in the objective function to explore the consensus information. Multi-manifold regularization is adopted for multi-view semi-supervised learning. Combination weights for all view in the norm regularization terms are adopted to exploit complementarity information among distinct views. Finally, an efficient alternating algorithm is proposed for optimization. Experiments are performed on various real world datasets, which give state-of-the-art generalization performance.

Using Domain Knowledge for Interpretable and Competitive Multi-Class Human Activity Recognition

Human activity recognition (HAR) has become an increasingly popular application of machine learning across a range of domains. Typically the HAR task that a machine learning algorithm is trained for requires separating multiple activities such as walking, running, sitting, and falling from each other. Despite a large body of work on multi-class HAR, and the well-known fact that the performance on a multi-class problem can be significantly affected by how it is decomposed into a set of binary problems, there has been little research into how the choice of multi-class decomposition method affects the performance of HAR systems. This paper presents the first empirical comparison of multi-class decomposition methods in a HAR context by estimating the performance of five machine learning algorithms when used in their multi-class formulation, with four popular multi-class decomposition methods, five expert hierarchies—nested dichotomies constructed from domain knowledge—or an ensemble of expert hierarchies on a 17-class HAR data-set which consists of features extracted from tri-axial accelerometer and gyroscope signals. We further compare performance on two binary classification problems, each based on the topmost dichotomy of an expert hierarchy. The results show that expert hierarchies can indeed compete with one-vs-all, both on the original multi-class problem and on a more general binary classification problem, such as that induced by an expert hierarchy’s topmost dichotomy. Finally, we show that an ensemble of expert hierarchies performs better than one-vs-all and comparably to one-vs-one, despite being of lower time and space complexity, on the multi-class problem, and outperforms all other multi-class decomposition methods on the two dichotomous problems.

ШКІЛЬНІ ТЕХНОЛОГІЇ ЗБЕРЕЖЕННЯ ЗДОРОВ’Я УЧНІВ: ЗАГАЛЬНА КЛАСИФІКАЦІЯ

The rapid loss of health of the younger generations during school education makes it a pressing problem for students to maintain their health during the learning process. This problem is extremely important not only for the general education environment but also for institutions of higher and postgraduate education, since the safety of the educational process for the health of students depends first and foremost on the teacher. However, to date, the problem of student health technologies has not been sufficiently studied. In particular, there is no single approach to the name and overall classification of such technologies. The article analyzes the views of scientists on the nature of student health technologies. Based on this, the term "school technologies of preserving students' health" and a general classification of student health preservation technologies are proposed for use, which logically integrates existing semantic interpretations. We believe that it is appropriate to separate school technologies for preserving students' health into appropriate health (technologies implemented in the educational process without compromising the health of the subjects: taking into account individual, age, gender characteristics; ergonomic, technical, organizational norms educational process, etc.) and health-saving (technologies aimed at shaping students' competences in maintaining their own health and that of other people: fostering a culture of health, health of the same lifestyle, sanogenic thinking etc.). It is noted that the results of theoretical analysis of the problem of general classification of school technologies of preserving students' health actualize the problem of further detailed classification of these technologies and ways of their introduction in the educational process of general educational institutions.

Mathematical model of the general problem of state classification in wireless sensor networks

This paper considers a mathematical model of the general classification problem in wireless sensor networks. We present the fields of application of sensor networks, the existing types of sensors used in data networks are listed, as well as the most typical conditions of sensor networks. The main attention is paid to the analysis of mathematical models of classification tasks, evaluating its parameters. As a result, there has been developed a mathematical model of the problem of the classification of states in wireless sensor networks, taking into account factors such as inaccuracies in the sensor readings, the dynamics of the number of sensors, dynamic network conditions, and their priority. This object allows you to move on to the next stage of the classification network conditions, the development of the classification algorithm.

General multi-view learning with maximum entropy discrimination

Maximum entropy discrimination (MED) is an excellent discriminative classification algorithm in view of maximum entropy and maximum margin principles, and can generate hard-margin support vector machines (SVMs) under certain condition. Recently, many versions of multi-view maximum entropy discrimination are proved to be effective in multi-view learning. However, those methods are designed only for the two-view classification problem. For example, alternative multi-view maximum entropy discrimination (AMvMED) and multi-kernel multi-view maximum entropy discrimination (MkMvMED) belong to the two-view classification method. In this paper, we extend AMvMED and MkMvMED to the general multi-view classification problems by jointly learning multiple different views in a non-pairwise way called general alternative multi-view maximum entropy discrimination (GAMvMED) and general multi-kernel multi-view maximum entropy discrimination (GM3ED). In GAMvMED, combination weight for the reconstruction of each view in regularization terms is learned to explore complementarity information among different views. Finally, we provide an efficient iteration algorithm with the classical convex quadratic programming for optimization. Simultaneously, we derive that GAMvMED and GM3ED are equivalent in form under the similar prior assumption. Experimental results confirm the effectiveness of our proposed methods.

A pattern recognition modeling approach based on the intelligent ensemble classifier: Application to identification and appraisal of water-flooded layers

Since the actual chromatogram data of the water-flooded layer have characteristics of multiple dimension, complexity and noise, it is difficult to accurately identify and appraise the water-flooded layer in the oil and gas reservoirs. Therefore, this article proposes a recognition modeling approach based on the intelligent ensemble classifier, integrated model-free Bayesian classifier, the AdaBoost algorithm and the support vector machine algorithm. The effective chromatogram characteristic information can be obtained using the curve fitting method. In order to transform the sparse classification problem into a general classification problem, the synthetic minority over-sampling technique algorithm is used to process an unbalanced training sample as a general training sample. Moreover, the model-free Bayesian classifier, AdaBoost and support vector machine algorithms are used as the base classifiers to train the ensemble classification model. Compared to the traditional single classification approach, the robustness and the effectiveness of the ensemble classifier model are validated through the standard data source from the UCI (University of California at Irvine) repository. Finally, the proposed model is applied in the identification and appraisal of the water-flooded layers in a complex oil and gas recognition system. The chromatogram characteristic information and the prediction results are obtained to provide more reliable water-flooded layer information, guide the process of reservoir exploration and improve the oil development efficiency.

Sensible functional linear discriminant analysis

Fisher’s linear discriminant analysis (LDA) is extended to both densely recorded functional data and sparsely observed longitudinal data for general c-category classification problems. An efficient approach is proposed to identify the optimal LDA projections in addition to managing the noninvertibility issue of the covariance operator emerging from this extension. To tackle the challenge of projecting sparse data to the LDA directions, a conditional expectation technique is employed. The asymptotic properties of the proposed estimators are investigated and asymptotically perfect classification is shown to be achievable in certain circumstances. The performance of this new approach is further demonstrated with both simulated data and real examples.

Robust video-based face recognition via M-estimator and image set collaborative representation

This paper presents a novel method for video-based face recognition (VFR) based on M-estimator and image set collaborative representation. Since a video is essentially an image set, the VFR problem can be cast as a special case of the image set-based face recognition (FR) problem. To measure the distance between the query image set and the gallery image set, we develop an M-estimator-based image set collaborative representation (MISCR) model. To implement MISCR, we devise an efficient half-quadratic-based optimization algorithm to tackle the complicated optimization problem. We also establish the convergence property of the devised algorithm. Our other contribution is to propose an MISCR-based classifier for the general image set classification problem, including VFR as a special case. The experiments using real-world benchmark databases demonstrate the efficacy and robustness of the proposed method for VFR.

Asymptotic distributions of kappa statistics and their differences with many raters, many rating categories and two conditions.

In clinical research and in more general classification problems, a frequent concern is the reliability of a rating system. In the absence of a gold standard, agreement may be considered as an indication of reliability. When dealing with categorical data, the well-known kappa statistic is often used to measure agreement. The aim of this paper is to obtain a theoretical result about the asymptotic distribution of the kappa statistic with multiple items, multiple raters, multiple conditions, and multiple rating categories (more than two), based on recent work. The result settles a long lasting quest for the asymptotic variance of the kappa statistic in this situation and allows for the construction of asymptotic confidence intervals. A recent application to clinical endoscopy and to the diagnosis of inflammatory bowel diseases (IBDs) is shortly presented to complement the theoretical perspective.

Hyperspectral Image Classification Using Principal Components-Based Smooth Ordering and Multiple 1-D Interpolation

This paper proposes a spectral-spatial classification algorithm based on principal components (PCs)-based smooth ordering and multiple 1-D interpolation, which can alleviate the general classification problems effectively. Because of the characteristics of hyperspectral image, there always exist easily separable samples (ESSs) and difficultly separable samples (DSSs) in view of the different sets of labeled samples. In this paper, the PC analysis is first used for reducing features and extracting the few first PCs of a hyperspectral image. Then, PC-based smooth ordering is designed for the separation of ESSs and DSSs, and multiple 1-D interpolation is used for the accurate classification of the ESSs. Next, the highly confident samples are selected from the ESSs by the spatial neighborhood information, which are added into the training set for the classification of DSSs. In the case of sufficient training samples, a supervised spectral-spatial method is used for classifying the DSSs by combining the spatial information built with popular extended multiattribute profiles. The proposed algorithm is compared with some state-of-the-art methods on three hyperspectral data sets. The results demonstrate that the presented algorithm achieves much better classification performance in terms of the accuracy and the computation time.

A FULLY AUTOMATED PIPELINE FOR CLASSIFICATION TASKS WITH AN APPLICATION TO REMOTE SENSING

Nowadays deep learning has been intensively in spotlight owing to its great victories at major competitions, which undeservedly pushed ‘shallow’ machine learning methods, relatively naive/handy algorithms commonly used by industrial engineers, to the background in spite of their facilities such as small requisite amount of time/dataset for training. We, with a practical point of view, utilized shallow learning algorithms to construct a learning pipeline such that operators can utilize machine learning without any special knowledge, expensive computation environment, and a large amount of labelled data. The proposed pipeline automates a whole classification process, namely feature-selection, weighting features and the selection of the most suitable classifier with optimized hyperparameters. The configuration facilitates particle swarm optimization, one of well-known metaheuristic algorithms for the sake of generally fast and fine optimization, which enables us not only to optimize (hyper)parameters but also to determine appropriate features/classifier to the problem, which has conventionally been a priori based on domain knowledge and remained untouched or dealt with naïve algorithms such as grid search. Through experiments with the MNIST and CIFAR-10 datasets, common datasets in computer vision field for character recognition and object recognition problems respectively, our automated learning approach provides high performance considering its simple setting (i.e. non-specialized setting depending on dataset), small amount of training data, and practical learning time. Moreover, compared to deep learning the performance stays robust without almost any modification even with a remote sensing object recognition problem, which in turn indicates that there is a high possibility that our approach contributes to general classification problems.

A FULLY AUTOMATED PIPELINE FOR CLASSIFICATION TASKS WITH AN APPLICATION TO REMOTE SENSING

Abstract. Nowadays deep learning has been intensively in spotlight owing to its great victories at major competitions, which undeservedly pushed ‘shallow’ machine learning methods, relatively naive/handy algorithms commonly used by industrial engineers, to the background in spite of their facilities such as small requisite amount of time/dataset for training. We, with a practical point of view, utilized shallow learning algorithms to construct a learning pipeline such that operators can utilize machine learning without any special knowledge, expensive computation environment, and a large amount of labelled data. The proposed pipeline automates a whole classification process, namely feature-selection, weighting features and the selection of the most suitable classifier with optimized hyperparameters. The configuration facilitates particle swarm optimization, one of well-known metaheuristic algorithms for the sake of generally fast and fine optimization, which enables us not only to optimize (hyper)parameters but also to determine appropriate features/classifier to the problem, which has conventionally been a priori based on domain knowledge and remained untouched or dealt with naïve algorithms such as grid search. Through experiments with the MNIST and CIFAR-10 datasets, common datasets in computer vision field for character recognition and object recognition problems respectively, our automated learning approach provides high performance considering its simple setting (i.e. non-specialized setting depending on dataset), small amount of training data, and practical learning time. Moreover, compared to deep learning the performance stays robust without almost any modification even with a remote sensing object recognition problem, which in turn indicates that there is a high possibility that our approach contributes to general classification problems.

An upper and lower CUSUM for signal normalization in the dendritic cell algorithm

Signal normalization is a part of signal formalization which is a vital data pre-processing constraint required for the functioning of the dendritic cell algorithm. In existing applications, most normalization algorithms are developed purposely for a specific application with drawing on human domain expertise and very few algorithms are designed for general problems. This makes it difficult for the inexperienced user to exploit existing approaches to another problem, particularly when the initial information about the problem is limited. Therefore, this study proposes a new signal normalization method for the dendritic cell algorithm that uses the statistical upper and lower cumulative sum so that the algorithm can be applied to general classification problems. In addition, a new method to calculate the anomaly threshold based on the average mature-contact antigen value is presented to suit the proposed algorithm. The proposed model is evaluated by applying it to eight universal classification datasets and assessing its performance according to four measurement metrics: detection rate, specificity, false alarm rate, and accuracy. Its performance is compared with that of the existing dendritic cell algorithm and four non-bio-inspired classifiers, namely, rough set, decision tree, naive Bayes, and multilayer perceptron. The results show that the proposed model outperforms the existing model and the other classifiers as well as demonstrates a significant improvement in terms of specificity, false alarm rate, and accuracy for all datasets. This indicates that the proposed normalization approach can be applied to general classification problems and can improve detection performance.

Towards generic image classification using tree-based learning: An extensive empirical study

This paper considers the general problem of image classification without using any prior knowledge about image classes. We study variants of a method based on supervised learning whose common steps are the extraction of random subwindows described by raw pixel intensity values and the use of ensemble of extremely randomized trees to directly classify images or to learn image features. The influence of method parameters and variants is thoroughly evaluated so as to provide baselines and guidelines for future studies. Detailed results are provided on 80 publicly available datasets that depict very diverse types of images (more than 3800 image classes and over 1.5 million images).

MetricFusion: Generalized metric swarm learning for similarity measure

Learning distance metrics for measuring the similarity between two data points in unsupervised and supervised pattern recognition has been widely studied in unconstrained face verification tasks. Motivated by the fact that enforcing single distance metric learning for verification via an empirical score threshold is not robust in uncontrolled experimental conditions, we therefore propose to obtain a metric swarm by learning local patches alike sub-metrics simultaneously that naturally formulates a generalized metric swarm learning (GMSL) model with a joint similarity score function solved by an efficient alternative optimization algorithm. Further, each sample pair is represented as a similarity vector via the well-learned metric swarm, such that the face verification task becomes a generalized SVM-alike classification problem. Therefore, the verification can be enforced in the represented metric swarm space that can well improve the robustness of verification under irregular data structure. Experiments are preliminarily conducted using several UCI benchmark datasets for solving general classification problem. Further, the face verification experiments on real-world LFW and PubFig datasets demonstrate that our proposed model outperforms several state-of-the-art metric learning methods.

The Types of Retinal Ganglion Cells: Current Status and Implications for Neuronal Classification

In the retina, photoreceptors pass visual information to interneurons, which process it and pass it to retinal ganglion cells (RGCs). Axons of RGCs then travel through the optic nerve, telling the rest of the brain all it will ever know about the visual world. Research over the past several decades has made clear that most RGCs are not merely light detectors, but rather feature detectors, which send a diverse set of parallel, highly processed images of the world on to higher centers. Here, we review progress in classification of RGCs by physiological, morphological, and molecular criteria, making a particular effort to distinguish those cell types that are definitive from those for which information is partial. We focus on the mouse, in which molecular and genetic methods are most advanced. We argue that there are around 30 RGC types and that we can now account for well over half of all RGCs. We also use RGCs to examine the general problem of neuronal classification, arguing that insights and methods from the retina can guide the classification enterprise in other brain regions.

,

Extreme Learning Machine (ELM) is a recently introduced learning algorithm for single hidden-layer feedforward neural network. Compared with classical learning algorithms in neural network, e.g. Back Propagation, ELM can achieve better performance with much shorter learning time. In the existing literature its better performance and comparison with Support Vector Machine (SVM), over regression and general classification problems, have caught the attention of many researchers. In this work, a comparison between ELM and SVM on identification of Erythemato-Squamous skin diseases is investigated. Detailed comparative studies were carried out through adequate experimentation. Experimental results indicated that ELM outperformed SVM. The effect of varying the size of training and testing sets on the performance of classifiers was also investigated in this study. The two techniques compared proved to be viable tools in this germane field of medical diagnosis.

A bag of cells approach for antinuclear antibodies HEp-2 image classification.

The antinuclear antibody (ANA) test via indirect immunofluorescence applied on Human Epithelial type 2 (HEp-2) cells is a pathology test commonly used to identify connective tissue diseases (CTDs). Despite its effectiveness, the test is still considered labor intensive and time consuming. Applying image-based computer aided diagnosis (CAD) systems is one of the possible ways to address these issues. Ideally, a CAD system should be able to classify ANA HEp-2 images taken by a camera fitted to a fluorescence microscope. Unfortunately, most prior works have primarily focused on the HEp-2 cell image classification problem which is one of the early essential steps in the system pipeline. In this work we directly tackle the specimen image classification problem. We aim to develop a system that can be easily scaled and has competitive accuracy. ANA HEp-2 images or ANA images are generally comprised of a number of cells. Patterns exhibiting in the cells are then used to make inference on the ANA image pattern. To that end, we adapted a popular approach for general image classification problems, namely a bag of visual words approach. Each specimen is considered as a visual document containing visual vocabularies represented by its cells. A specimen image is then represented by a histogram of visual vocabulary occurrences. We name this approach as the Bag of Cells approach. We studied the performance of the proposed approach on a set of images taken from 262 ANA positive patient sera. The results show the proposed approach has competitive performance compared to the recent state-of-the-art approaches. Our proposal can also be expanded to other tests involving examining patterns of human cells to make inferences.

A new approach for imbalanced data classification based on data gravitation

Imbalanced classification is an important machine learning research topic that troubles most general classification models because of the imbalanced class distribution. A newly developed physical-inspired classification method, i.e., the data gravitation-based classification (DGC) model, performs well in many general classification problems. However, like other general classifiers, the performance of DGC suffers in imbalanced tasks. In this study, we develop a specific DGC model namely Imbalanced DGC (IDGC) model for imbalanced problems. The amplified gravitation coefficient (AGC) is introduced for gravitation computing. AGC is a type of coefficient that contains class imbalance information, which can strengthen and weaken the gravitational field of the minority and majority classes. We also design a fitness evaluation function in the weight optimization procedure of the data distribution to ensure that the model parameters adapt to the imbalanced class distributions. A total of 44 binary class data sets and 15 multiclass imbalanced data sets are used to test the performance of the proposed method. Experimental results show that the adapted DGC model is effective for imbalanced problems.