One-class SVM Classifier Research Articles

Rapid Eye Movement Sleep Behavior Disorder as an Outlier Detection Problem

Idiopathic rapid eye movement (REM) sleep behavior disorder is a strong early marker of Parkinson's disease and is characterized by REM sleep without atonia and/or dream enactment. Because these measures are subject to individual interpretation, there is consequently need for quantitative methods to establish objective criteria. This study proposes a semiautomatic algorithm for the early detection of Parkinson's disease. This is achieved by distinguishing between normal REM sleep and REM sleep without atonia by considering muscle activity as an outlier detection problem. Sixteen healthy control subjects, 16 subjects with idiopathic REM sleep behavior disorder, and 16 subjects with periodic limb movement disorder were enrolled. Different combinations of five surface electromyographic channels, including the EOG, were tested. A muscle activity score was automatically computed from manual scored REM sleep. This was accomplished by the use of subject-specific features combined with an outlier detector (one-class support vector machine classifier). It was possible to correctly separate idiopathic REM sleep behavior disorder subjects from healthy control subjects and periodic limb movement subjects with an average validation area under the receiver operating characteristic curve of 0.993 when combining the anterior tibialis with submentalis. Additionally, it was possible to separate all subjects correctly when the final algorithm was tested on 12 unseen subjects. Detection of idiopathic REM sleep behavior disorder can be regarded as an outlier problem. Additionally, the EOG channels can be used to detect REM sleep without atonia and is discriminative better than the traditional submentalis. Furthermore, based on data and methodology, arousals and periodic limb movements did only have a minor influence on the quantification of the muscle activity. Analysis of muscle activity during nonrapid eye movement sleep may improve the separation even further.

A One-Class SVM Based Tool for Machine Learning Novelty Detection in HVAC Chiller Systems

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) chiller systems can lead to discomfort for the occupants, energy wastage, unreliability and shorter equipment life. Such faults need to be detected early to prevent further escalation and energy losses. Commonly, data regarding unforeseen phenomena and abnormalities are rare or are not available at the moment for HVAC installations: for this reason in this paper an unsupervised One-Class SVM classifier employed as a novelty detection system to identify unknown status and possible faults is presented. The approach, that exploits Principal Component Analysis to accent novelties w.r.t. normal operations variability, has been tested on a HVAC literature dataset.

Predicting complications of percutaneous coronary intervention using a novel support vector method

To explore the feasibility of a novel approach using an augmented one-class learning algorithm to model in-laboratory complications of percutaneous coronary intervention (PCI). Data from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) multicenter registry for the years 2007 and 2008 (n=41 016) were used to train models to predict 13 different in-laboratory PCI complications using a novel one-plus-class support vector machine (OP-SVM) algorithm. The performance of these models in terms of discrimination and calibration was compared to the performance of models trained using the following classification algorithms on BMC2 data from 2009 (n=20 289): logistic regression (LR), one-class support vector machine classification (OC-SVM), and two-class support vector machine classification (TC-SVM). For the OP-SVM and TC-SVM approaches, variants of the algorithms with cost-sensitive weighting were also considered. The OP-SVM algorithm and its cost-sensitive variant achieved the highest area under the receiver operating characteristic curve for the majority of the PCI complications studied (eight cases). Similar improvements were observed for the Hosmer-Lemeshow χ(2) value (seven cases) and the mean cross-entropy error (eight cases). The OP-SVM algorithm based on an augmented one-class learning problem improved discrimination and calibration across different PCI complications relative to LR and traditional support vector machine classification. Such an approach may have value in a broader range of clinical domains.

Supervised Expert System for Wearable MEMS Accelerometer-Based Fall Detector

Falling is one of the main causes of trauma, disability, and death among older people. Inertial sensors-based devices are able to detect falls in controlled environments. Often this kind of solution presents poor performances in real conditions. The aim of this work is the development of a computationally low-cost algorithm for feature extraction and the implementation of a machine-learning scheme for people fall detection, by using a triaxial MEMS wearable wireless accelerometer. The proposed approach allows to generalize the detection of fall events in several practical conditions. It appears invariant to the age, weight, height of people, and to the relative positioning area (even in the upper part of the waist), overcoming the drawbacks of well-known threshold-based approaches in which several parameters need to be manually estimated according to the specific features of the end user. In order to limit the workload, the specific study on posture analysis has been avoided, and a polynomial kernel function is used while maintaining high performances in terms of specificity and sensitivity. The supervised clustering step is achieved by implementing an one-class support vector machine classifier in a stand-alone PC.

Kernel-Based Mapping of Orographic Rainfall Enhancement in the Swiss Alps as Detected by Weather Radar

In this paper, we develop a data-driven methodology to characterize the likelihood of orographic precipitation enhancement using sequences of weather radar images and a digital elevation model (DEM). Geographical locations with topographic characteristics favorable to enforce repeatable and persistent orographic precipitation such as stationary cells, upslope rainfall enhancement, and repeated convective initiation are detected by analyzing the spatial distribution of a set of precipitation cells extracted from radar imagery. Topographic features such as terrain convexity and gradients computed from the DEM at multiple spatial scales as well as velocity fields estimated from sequences of weather radar images are used as explanatory factors to describe the occurrence of localized precipitation enhancement. The latter is represented as a binary process by defining a threshold on the number of cell occurrences at particular locations. Both two-class and one-class support vector machine classifiers are tested to separate the presumed orographic cells from the nonorographic ones in the space of contributing topographic and flow features. Site-based validation is carried out to estimate realistic generalization skills of the obtained spatial prediction models. Due to the high class separability, the decision function of the classifiers can be interpreted as a likelihood or susceptibility of orographic precipitation enhancement. The developed approach can serve as a basis for refining radar-based quantitative precipitation estimates and short-term forecasts or for generating stochastic precipitation ensembles conditioned on the local topography.

Active SVM-based relevance feedback using multiple classifiers ensemble and features reweighting

Relevance feedback (RF) is an effective approach to bridge the gap between low-level visual features and high-level semantic meanings in content-based image retrieval (CBIR). The active support vector machine (SVM) based RFs have been popular because they can outperform many other classifiers when the size of the training set is small, but they are often very complex and some unsatisfactory relevance of results occur frequently. To overcome the above limitations, an active SVM-based RF using multiple classifiers ensemble and features reweighting is proposed in this paper. Firstly, we select the most informative images by using active learning method for user to label, and quickly learn a boundary that separates the images that satisfy the user's query concept from the rest of the dataset. Secondly, the feature space is modified dynamically by appropriately weighting the descriptive features according to a set of statistical characteristics. Then, a set of moderate accurate one-class SVM classifiers are trained separately by using different new sub-features vectors. Finally, we compute the weight vector of component SVM classifiers dynamically by using the parameters for positive and negative samples, and combine the results of the component classifiers to form an output code as a hypothesized solution to the overall image retrieval problem. Extensive simulations on large databases show that the proposed algorithm is significantly more effective than the state-of-the-art approaches.

Semisupervised One-Class Support Vector Machines for Classification of Remote Sensing Data

This paper presents two semisupervised one-class support vector machine (OC-SVM) classifiers for remote sensing applications. In <emphasis emphasistype="boldital" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">one-class</emphasis> image classification, one tries to detect pixels belonging to one of the classes in the image and reject the others. When few labeled pixels of only one class are available, obtaining a reliable classifier is a difficult task. In the particular case of SVM-based classifiers, this task is even harder because the free parameters of the model need to be finely adjusted, but no clear criterion can be adopted. In order to improve the OC-SVM classifier accuracy and alleviate the problem of free-parameter selection, the information provided by unlabeled samples present in the scene can be used. In this paper, we present two state-of-the-art algorithms for semisupervised one-class classification for remote sensing classification problems. The first proposed algorithm is based on modifying the OC-SVM kernel by modeling the data marginal distribution with the graph Laplacian built with both labeled and unlabeled samples. The second one is based on a simple modification of the standard SVM cost function which penalizes more the errors made when classifying samples of the target class. The good performance of the proposed methods is illustrated in four challenging remote sensing image classification scenarios where the goal is to detect one of the classes present on the scene. In particular, we present results for multisource urban monitoring, hyperspectral crop detection, multispectral cloud screening, and change-detection problems. Experimental results show the suitability of the proposed techniques, particularly in cases with few or poorly representative labeled samples.

Urban building damage detection from very high resolution imagery using OCSVM and spatial features

The availability of commercial very high resolution (VHR) satellite imagery makes it possible to detect and assess building damage in the aftermath of earthquake disasters using these data. Although conventional change detection methods may be used to assess the building damage, the analysis is directed to all classes, both damaged and undamaged, but is not focused on the class of interest. In this paper, we proposed to detect the building damage in urban environments from multitemporal VHR image data using the One-Class Support Vector Machine (OCSVM), a recently developed one-class classifier, which requires training samples from the building damage only. This was illustrated with building damage detection in an urban environment from multitemporal Quickbird images. The detection was conducted at pixel level and object level. Different input vectors for the OCSVM classifier were tested in order to assess the discrimination power of spectral and spatial features: pixel level spectral features and texture features, as well as object‐based features. The results showed that the OCSVM performed better on the object level, with an overall accuracy of 82.33% and a kappa coefficient of 60.09%. The results also showed that the OCSVM provides a useful framework that can combine different features and focus on the building damage class of interest. More spatial features are needed to be exploited to obtain more accurate detection results in future studies.

Ensemble one-class support vector machines for content-based image retrieval

In order to narrow semantic gap between user query concept and low-level features in content-based image retrieval, SVM-based relevance feedback techniques are developed to learn user’s query concept by labeling some samples. The major difficulty in relevance feedback is to estimate the support of target image in high-dimensional feature space with small number of training samples. To overcome this limitation, we propose an ensemble method to boost image retrieval accuracy and to improve its generalization performance. Images are segmented into multiple instances. A set of moderate accurate one-class support vector machine classifiers are trained separately by using different sub-features extracted from instances. The ensemble method results in a highly accurate by combining moderately accurate weak classifiers. Our propose ensemble scheme not only provides a robust mechanism in selecting strong query concept related images for relevant feedback, but also achieves a generalization performance in image retrieval.

A Method for Locating Digital Evidences with Outlier Detection Using Support Vector Machine

One of the biggest challenges facing digital investigators is the sheer volume of data that must be searched in locating the digital evidence. How to efficiently locate the evidence relating to the computer crime while maintaining accuracy is becoming a research focus. In this paper, we introduce a two-tier method to automate the process of locating the digital evidence, which first employ a one-class Support Vector Machine (SVM) outlier detector to filter out insignificant records for forensic investigators and then use a group of one-class SVM classifiers (trained with the expert knowledge or interested samples for an investigator based on a different feature vector) to further analyze the output of the outlier detector to improve the accuracy of investigation. The effectiveness of the proposed method for locating digital evidence is demonstrated using the public datasets: KDD Cup99 (Knowledge Discovery and Data-mining) intrusion detection dataset.

Prediction of human microRNA hairpins using only positive sample learning

MicroRNAs (miRNAs) are small molecular non-coding RNAs that have important roles in the post-transcriptional mechanism of animals and plants. They are commonly 21-25 nucleo- tides (nt) long and derived from 60-90 nt RNA hairpin structures, called miRNA hairpins. A lar- ger number of sequence segments in the human genome have been computationally identified with such 60-90 nt hairpins, however the major- ity of them are not miRNA hairpins. Most exist- ing computational methods for predicting miRNA hairpins are based on a two-class classi- fier to distinguish between miRNA hairpins and other sequence segments with hairpin struc- tures. The difficulty of these methods is how to select hairpins as negative examples of miRNA hairpins in the training dataset, since only a few miRNA hairpins are available. Therefore, these classifiers may be mis-trained due to some false negative examples of the training dataset. In this paper, we introduce a one-class support vector machine (SVM) method to predict miRNA hair- pins among the hairpin structures. Different from existing methods for predicting miRNA hairpins, the one-class SVM classifier is trained only on the information of the miRNA class. We also il- lustrate some examples of predicting miRNA hairpins in human chromosomes 10, 15, and 21, where our method overcomes the above disad- vantages of existing two-class methods.

Kernel Fisher Discriminants for Outlier Detection

The problem of detecting atypical objects or outliers is one of the classical topics in (robust) statistics. Recently, it has been proposed to address this problem by means of one-class SVM classifiers. The method presented in this letter bridges the gap between kernelized one-class classification and gaussian density estimation in the induced feature space. Having established the exact relation between the two concepts, it is now possible to identify atypical objects by quantifying their deviations from the gaussian model. This model-based formalization of outliers overcomes the main conceptual shortcoming of most one-class approaches, which, in a strict sense, are unable to detect outliers, since the expected fraction of outliers has to be specified in advance. In order to overcome the inherent model selection problem of unsupervised kernel methods, a cross-validated likelihood criterion for selecting all free model parameters is applied. Experiments for detecting atypical objects in image databases effectively demonstrate the applicability of the proposed method in real-world scenarios.

Kernel Fisher Discriminants for Outlier Detection

The problem of detecting atypical objects or outliers is one of the classical topics in (robust) statistics. Recently, it has been proposed to address this problem by means of one-class SVM classifiers. The method presented in this letter bridges the gap between kernelized one-class classification and gaussian density estimation in the induced feature space. Having established the exact relation between the two concepts, it is now possible to identify atypical objects by quantifying their deviations from the gaussian model. This model-based formalization of outliers overcomes the main conceptual shortcoming of most one-class approaches, which, in a strict sense, are unable to detect outliers, since the expected fraction of outliers has to be specified in advance. In order to overcome the inherent model selection problem of unsupervised kernel methods, a cross-validated likelihood criterion for selecting all free model parameters is applied. Experiments for detecting atypical objects in image databases effectively demonstrate the applicability of the proposed method in real-world scenarios.