EEG signal classification using universum support vector machine

This paper analyzes the underlying complexity and non-linearity of electroencephalogram (EEG) signals by computing a novel multi-scale entropy measure for the classification of seizure, seizure-free and normal EEG signals. The quality factor (Q) based multi-scale entropy measure is proposed to compute the entropy of the EEG signal in different frequency-bands of interest. The Q -based entropy (QEn) is computed by decomposing the signal with the tunable-Q wavelet transform (TQWT) into the number of sub-bands and estimating K-nearest neighbor (K-NN) entropies from various sub-bands cumulatively. The optimal selection of Q and the redundancy parameter (R) of TQWT showed better robustness for entropy computation in the presence of high- and low-frequency components. The extracted features are fed to the support vector machine (SVM) classifier with the wrapper-based feature selection method. The proposed method has achieved accuracy of 100% in classifying normal (eyes-open and eyes-closed) and seizure EEG signals, 99.5% in classifying seizure-free EEG signals (from the hippocampal formation of the opposite hemisphere of the brain) from seizure EEG signals and 98% in classifying seizure-free EEG signals (from the epileptogenic zone) from seizure EEG signals, respectively, using the SVM classifier. We have also achieved classification accuracies of 99% and 98.6% in classifying seizure versus non-seizure EEG signals and the individual three classes, namely normal, seizure-free and seizure EEG signals, respectively. The performance measure of the proposed multi-scale entropy has been found to be comparable with the existing state of the art epileptic EEG signals classification methods studied using the same database.

abstractUniversum, a third class not belonging to either class of the classification problem, allows to incorporate the prior knowledge into the learning process. A lot of previous work have demonstrated that the Universum is helpful to the supervised and semi-supervised classification. Moreover, Universum has already been introduced into the support vector machine (SVM) and twin support vector machine (TSVM) to enhance the generalisation performance. To further increase the generalisation performance, we propose a least squares TSVM with Universum data (-TSVM) in this paper. Our -TSVM possesses the following advantages: first, it exploits Universum data to improve generalisation performance. Besides, it implements the structural risk minimisation principle by adding a regularisation to the objective function. Finally, it costs less computing time by solving two small-sized systems of linear equations instead of a single larger-sized quadratic programming problem. To verify the validity of our proposed algorithm, we conduct various experiments around the size of labelled samples and the number of Universum data on data-sets including seven benchmark data-sets, Toy data, MNIST and Face images. Empirical experiments indicate that Universum contributes to making prediction accuracy improved even stable. Especially when fewer labelled samples given, -TSVM is far superior to the improved LS-TSVM (ILS-TSVM), and slightly superior to the -TSVM.

KNN weighted reduced universum twin SVM for class imbalance learning

Epilepsy is a neurological disorder that affects people of any age, which can be detected by Electroencephalogram (EEG) signals. This paper proposes a novel method called Volume of Phase Space Representation (VOPSR) to classify seizure and seizure-free EEG signals automatically. Primarily, the recorded EEG signal is disintegrated into several Intrinsic Mode Functions (IMFs) using the Empirical Mode Decomposition (EMD) method and the three-dimensional phase space have been reconstructed for the obtained IMFs. The volume is measured for the obtained 3D-PSR for different IMFs called VOPSR, which is used as a feature set for the classification of Epileptic seizure EEG signals. Support vector machine (SVM) is used as a classifier for the classification of epileptic and epileptic-free EEG signals. The classification performance of the proposed method is evaluated under different kernels such as Linear, Polynomial and Radial Basis Function (RBF) kernels. Finally, the proposed method outperforms noteworthy state-of-the-art classification methods in the context of epileptic EEG signals, achieving 99.13% accuracy (average) with the Linear, Polynomial, and RBF kernels. The proposed technique can be used to detect epilepsy from the EEG signals automatically without human intervention.

An improved non-parallel Universum support vector machine and its safe sample screening rule

Epilepsy is the most common neurological disorder characterized by recurrent seizures due to the abnormality of brain neuronal discharge. One of the tools to help neurologists diagnose epilepsy relies on EEG signals extensively because it can represent the brain's neuron activity. Due to the EEG signal being considered a complex biological signal, the signal complexity-based feature extraction method is suitable for characterizing epileptic EEG signals. One method for extracting EEG signals is entropy which is able to quantify signal complexity. Several previously proposed methods were wavelet entropy and IMF entropy. This study proposed multi-distance signal level difference (MSLD) entropy as a feature extraction method to extract the EEG signal complexity. The decomposed signal from MSLD was then extracted using entropy measurement. In the classification process, a Support Vector Machine (SVM) was applied to classify the normal, pre-ictal, and ictal signals of EEG signals. The highest accuracy of 96% was obtained using cubic SVM as the classifier. The proposed method provides an alternative method for analyzing EEG or other biomedical signals based on signal complexity.

Event Abstract Back to Event High Performance EEG Signal Classification using Classifiability and the Twin SVM Sumit Soman1* and Jayadeva .1 1 Indian Institute of Technology, Delhi, India, Department of Electrical Engineering, India Classification of Electroencephalogram (EEG) data for imagined motor movements has been a challenge in the design and development of Brain Computer Interfaces (BCIs). There are two principle challenges. The first is the variability in the recorded EEG data, which manifests across trials as well as across individuals. Consequently, features that are more discriminative need to be identified before any pattern recognition technique can be applied. The second challenge is in the pattern recognition domain. The number of data samples in a class of interest, e.g. a specific action, is a small fraction of the total data, which is composed of samples corresponding to all actions of all users. Building a robust classifier when learning from a highly unbalanced dataset is very difficult; minimizing the classification error typically causes the larger class to overwhelm the smaller one. We show that the combination of ‘classifiability’ for selecting the optimal frequency band and the use of the Twin Support Vector Machine (Twin SVM) for classification, yields significantly improved generalization. On benchmark BCI Competition datasets, the proposed approach often yields up to 20% improvement over the state-of-the-art. Figure 1 References Soman, Sumit and Jayadeva. "High performance EEG signal classification using classifiability and the Twin SVM." Applied Soft Computing Vol. 30 (May 2015): pp. 305-318 [http://www.sciencedirect.com/science/article/pii/S1568494615000204] Keywords: Brain Computer Interface, Classifiability, Support Vector Machines, Motor Imagery (MI), multiclass classification, Unbalanced data Conference: German-Japanese Adaptive BCI Workshop, Kyoto, Japan, 28 Oct - 29 Oct, 2015. Presentation Type: Poster presentation Topic: Adaptive BCI Citation: Soman S and . J (2015). High Performance EEG Signal Classification using Classifiability and the Twin SVM. Front. Comput. Neurosci. Conference Abstract: German-Japanese Adaptive BCI Workshop. doi: 10.3389/conf.fncom.2015.56.00020 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 18 Sep 2015; Published Online: 04 Nov 2015. * Correspondence: Mr. Sumit Soman, Indian Institute of Technology, Delhi, India, Department of Electrical Engineering, Delhi, India, sumit.soman@gmail.com Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Sumit Soman Jayadeva . Google Sumit Soman Jayadeva . Google Scholar Sumit Soman Jayadeva . PubMed Sumit Soman Jayadeva . Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Epileptic seizure is the common neurological disorder, which is generally identified by electroencephalogram (EEG) signals. In this paper, a new feature extraction methodology based on optimum allocation sampling (OAS) and Teager energy operator (TEO) is proposed for detection of seizure EEG signals. The OAS scheme selects the finite length homogeneous sequence from non-homogeneous recorded EEG signal. The trend of selected sequence by OAS is still non-linear, which is analyzed by non-linear operator TEO. The TEO convert non-linear but homogenous EEG sequence into amplitude-frequency modulated (AM-FM) components. The statistical measures of AM-FM components used as input features to least squares support vector machine classifier for classification of seizure and seizure-free EEG signals. The proposed methodology is evaluated on a benchmark epileptic seizure EEG database. The experimental results demonstrate that the proposed scheme has capability to effectively distinguish seizure and seizure-free EEG signals.

Twin support vector machine (TWSVM) is a new machine learning method, as opposed to solving a single quadratic programming problem in support vector machine (SVM), which generates two nonparallel hyperplanes by solving two smaller size quadratic programming problems. However, the TWSVM obtains the final classifier by giving the same importance to all training samples which may be important for classification performance. In order to address this problem, in this paper, we propose a novel entropy-based fuzzy twin bounded support vector machine (EFTBSVM) for binary classification problems. By considering the fuzzy membership value for each sample and assigning it based on the entropy value, the samples with higher class certainty are assigned to relatively larger fuzzy membership. In addition, the proposed EFTBSVM not only maintains the superior characteristics of the TWSVM but also exploits the structural risk minimization principle by introducing a regularization term. The experimental results achieved on synthetic datasets and benchmark datasets illustrate the effectiveness of the proposed method.

A novel ?-twin support vector machine with Universum data (U?$\mathfrak {U}_{\nu }$-TSVM) is proposed in this paper. U?$\mathfrak {U}_{\nu }$-TSVM allows to incorporate the prior knowledge embedded in the unlabeled samples into the supervised learning. It aims to utilize these prior knowledge to improve the generalization performance. Different from the conventional U$\mathfrak {U}$-SVM, U?$\mathfrak {U}_{\nu }$-TSVM employs two Hinge loss functions to make the Universum data lie in a nonparallel insensitive loss tube, which makes it exploit these prior knowledge more flexibly. In addition, the newly introduced parameters ?1, ?2 in the U?$\mathfrak {U}_{\nu }$-TSVM have better theoretical interpretation than the penalty factor c in the U$\mathfrak {U}$-TSVM. Numerical experiments on seventeen benchmark datasets, handwritten digit recognition, and gender classification indicate that the Universum indeed contributes to improving the prediction accuracy. Moreover, our U?$\mathfrak {U}_{\nu }$-TSVM is far superior to the other three algorithms (U$\mathfrak {U}$-SVM, ?-TSVM and U$\mathfrak {U}$-TSVM) from the prediction accuracy.

Truncated loss functions are robust to class noise and outliers. A robust twin bounded support vector machine is proposed in this paper that truncates the growth of its loss functions at a pre-specified point, thus, flattens the function that pre-specified score afterwards. Moreover, to make the proposed method capable of handling datasets with different imbalance ratio, cost-sensitive learning is implemented by scaling the total error of the classes based on the number of samples of each class. However, the adopted loss functions take a non-convex structure which does not always assure global optimum. To handle this issue, we suggest concave-convex procedure (CCCP) to ensure global convergence by decomposing the cost functions into additions of one convex and one concave part. The behaviour of the proposed method for varied imbalance ratio is analysed experimentally and depicted graphically. Classification performance of the proposed method in terms of AUC, F-Measure and G-Mean is compared with other related methods, viz. Hinge loss support vector machine (SVM), Ramp loss SVM (RSVM), twin SVM (TWSVM), twin bounded SVM (TBSVM), Pinball loss SVM (pin-SVM), entropy-based fuzzy SVM (EFSVM), non-parallel hyperplane Universum SVM (U-NHSVM), stochastic gradient twin support vector machine (SGTSVM), k-nearest neighbor (KNN)-based maximum margin and minimum volume hyper-sphere machine (KNN-M3VHM) and affinity and class probability-based fuzzy SVM (ACFSVM) on several real-world datasets with imbalance ratio raging from low to high. Further, to establish the significance of the proposed method in pattern classification, pair-wise statistical comparison of the methods is performed based on their average ranks on AUC. Experimental results are convincing.

Over the past decades, 1-norm techniques based on algorithms are widely used to suppress input features. Quite different from traditional 1-norm support vector machine (SVM), direct 1-norm optimization based on the primal problem of nonparallel plane classifiers like generalized proximal support vector machine, twin support vector machine (TWSVM) and least squares twin support vector machine (LSTSVM) are not capable of generating very sparse solutions that are vital for classification and can make them easier to store and faster to compute. To address the issue, in this paper, we develop a feature selection method for LSTSVM, called a feature selection method for nonparallel plane support vector machine classification (FLSTSVM), which is specially designed for strong feature suppression. We incorporate a Tikhonov regularization term to the objective of LSTSVM, and then minimize its 1-norm measure. Solution of FLSTSVM can follow directly from solving two smaller quadratic programming problems (QPPs) arising from two primal QPPs as opposed to two dual ones in TWSVM. FLSTSVM is capable of generating very sparse solutions. This means that FLSTSVM can reduce input features, for the linear case. When a nonlinear classifier is used, few kernel functions determine the classifier. In addition to having strong feature suppression, the edge of our method still lies in its faster computing time compared to that of TWSVM, Newton Method for Linear Programming SVM (NLPSVM) and LPNewton. Lastly, this algorithm is compared on public data sets, as well as an Exclusive Or (XOR) example.

Classical support vector machine (SVM) and its twin variant twin support vector machine (TWSVM) utilize the Hinge loss that shows linear behaviour, whereas the least squares version of SVM (LSSVM) and twin least squares support vector machine (LSTSVM) uses L2-norm of error which shows quadratic growth. The robust Huber loss function is considered as the generalization of Hinge loss and L2-norm loss that behaves like the quadratic L2-norm loss for closer error points and the linear Hinge loss after a specified distance. Three functional iterative approaches based on generalized Huber loss function are proposed in this paper to solve support vector classification problems of which one is based on SVM, i.e. generalized Huber support vector machine and the other two are in the spirit of TWSVM, namely generalized Huber twin support vector machine and regularization on generalized Huber twin support vector machine. The proposed approaches iteratively find the solutions and eliminate the requirements to solve any quadratic programming problem (QPP) as for SVM and TWSVM. The main advantages of the proposed approach are: firstly, utilize the robust Huber loss function for better generalization and for lesser sensitivity towards noise and outliers as compared to quadratic loss; secondly, it uses functional iterative scheme to find the solution that eliminates the need to solving QPP and also makes the proposed approaches faster. The efficacy of the proposed approach is established by performing numerical experiments on several real-world datasets and comparing the result with related methods, viz. SVM, TWSVM, LSSVM and LSTSVM. The classification results are convincing.

Support vector machine (SVM) has attracted substantial interest in the community of machine learning. As the extension of SVM, nonparallel hyperplane SVM (NHSVM) classification algorithms have become current researching hot spots in machine learning during the last few years. For binary classification tasks, the idea of NHSVM algorithms is to find a hyperplane for each class, such that each hyperplane is proximal to the data points of one class and far from the data points of the other class. Compared with the classical SVM, NHSVM algorithms have lower computational complexity, work better on XOR problems and can get better generalization performance. This paper reviews three representative NHSVM algorithms, including generalized eigenvalue proximal SVM (GEPSVM), twin SVM (TWSVM) and projection twin SVM (PTSVM), and gives the research progress of them. The aim of this overview is to provide an insightful organization of current developments of NHSVM algorithms, identify their limitations and give suggestions for further research.

In the field of machine learning, support vector machine is a supervised learning model which can analyze the data and identify patterns, and this theory is used for classification and regression analysis which is related to the learning algorithm.Support vector machine is a new machine learning method based on statistical learning theory, and has become a hot research topic in the machine learning field because of its excellent learning performance.Support vector machine is also a learning machine based on kernel function, and its generalization ability depends on the chosen kernel function to a great extent.Nevertheless, traditional support vector machine cannot achieve the desired results when training large-scale data set.Therefore, in order to improve training efficiency and generalization performance of support vector machine, we have to improve this algorithm.Although support vector machine has great advantages in theory, the research on its application is comparatively delayed.In this paper, we focus on novel support vector machine generated from multi-subjects amalgamation, such as fuzzy support vector machine, granular support vector machine, twin support vector machine.We analyze a large number of literature review about development trend of support vector machine and its applications in computer field, laying foundation for the systematic research on the correlation algorithm based on support vector machine.