Utilizing Support Vector Machine Research Articles

A data mining approach to the diagnosis of failure modes for two serial fastened sandwich composite plates

The aim of this study is to diagnose and classify the failure modes for two serial fastened sandwich composite plates using data mining techniques. The composite material used in the study was manufactured using glass fiber reinforced layer and aluminum sheets. Obtained results of previous experimental study for sandwich composite plates, which were mechanically fastened with two serial pins or bolts were used for classification of failure modes. Furthermore, experimental data from previous study consists of different geometrical parameters for various applied preload moments as 0 (pinned), 2, 3, 4, and 5 Nm (bolted). In this study, data mining methods were applied by using these geometrical parameters and pinned/bolted joint configurations. Therefore, three geometrical parameters and 100 test data were used for classification by utilizing support vector machine, Naive Bayes, K-Nearest Neighbors, Logistic Regression, and Random Forest methods. According to experiments, Random Forest method achieved better results than others and it was appropriate for diagnosing and classification of the failure modes. Performances of all data mining methods used were discussed in terms of accuracy and error ratios.

Prediction of global solar radiation using support vector machines

ABSTRACTThis article utilizes Support Vector Machines (SVM) for predicting global solar radiation (GSR) for Sharurha, a city in the southwest of Saudi Arabia. The SVM model was trained using measured air temperature and relative humidity. Measured data of 1812 values for the period from 1998–2002 were obtained. The measurement data of 1600 were used for training the SVM, and the remaining 212 were used for comparison between the measured and predicted values of GSR. The GSR values were predicted using the following four combinations of data sets: (i) Daily mean air temperature and day of the year as inputs, and global solar radiation as output; (ii) daily maximum air temperature and day of the year as inputs, and GSR as output; (iii) daily mean air temperature and relative humidity and day of the year as inputs, and GSR as output; and (iv) daily mean air temperature, day of the year, relative humidity, and previous day’s GSR as inputs, and GSR as output. The mean square error was found to be 0.0027, 0.0023, 0.0021, and 7.65 × 10−4 for case (i), (ii,), (iii), and (iv) respectively, while the corresponding absolute mean percentage errors were 5.64, 5.08, 4.48, and 2.8%. Obtained results show that the SVM method is capable of predicting GSR from measured values of temperature and relative humidity.

A support vector machine model provides an accurate transcript-level-based diagnostic for major depressive disorder.

Major depressive disorder (MDD) is a critical cause of morbidity and disability with an economic cost of hundreds of billions of dollars each year, necessitating more effective treatment strategies and novel approaches to translational research. A notable barrier in addressing this public health threat involves reliable identification of the disorder, as many affected individuals remain undiagnosed or misdiagnosed. An objective blood-based diagnostic test using transcript levels of a panel of markers would provide an invaluable tool for MDD as the infrastructure—including equipment, trained personnel, billing, and governmental approval—for similar tests is well established in clinics worldwide. Here we present a supervised classification model utilizing support vector machines (SVMs) for the analysis of transcriptomic data readily obtained from a peripheral blood specimen. The model was trained on data from subjects with MDD (n=32) and age- and gender-matched controls (n=32). This SVM model provides a cross-validated sensitivity and specificity of 90.6% for the diagnosis of MDD using a panel of 10 transcripts. We applied a logistic equation on the SVM model and quantified a likelihood of depression score. This score gives the probability of a MDD diagnosis and allows the tuning of specificity and sensitivity for individual patients to bring personalized medicine closer in psychiatry.

A Robust Intrusion Detection System by Utilizing Support Vector Machine and Error Back Propagation Neural Network

A Robust Intrusion Detection System by Utilizing Support Vector Machine and Error Back Propagation Neural Network

Health monitoring of floating structure by utilizing support vector machines

Health monitoring of floating structure by utilizing support vector machines

An Efficient User Verification System Using Angle-Based Mouse Movement Biometrics

Biometric authentication verifies a user based on its inherent, unique characteristics—who you are. In addition to physiological biometrics, behavioral biometrics has proven very useful in authenticating a user. Mouse dynamics, with their unique patterns of mouse movements, is one such behavioral biometric. In this article, we present a user verification system using mouse dynamics, which is transparent to users and can be naturally applied for continuous reauthentication. The key feature of our system lies in using much more fine-grained (point-by-point) angle-based metrics of mouse movements for user verification. These new metrics are relatively unique from person to person and independent of a computing platform. Moreover, we utilize support vector machines (SVMs) for quick and accurate classification. Our technique is robust across different operating platforms, and no specialized hardware is required. The efficacy of our approach is validated through a series of experiments, which are based on three sets of user mouse movement data collected in controllable environments and in the field. Our experimental results show that the proposed system can verify a user in an accurate and timely manner, with minor induced system overhead.

Speech Emotion Recognition based on SVM as Both Feature Selector and Classifier

The aim of this paper is to utilize Support Vector Machine (SVM) as feature selection and classification techniques for audio signals to identify human emotional states. One of the major bottlenecks of common speech emotion recognition techniques is to use a huge number of features per utterance which could significantly slow down the learning process, and it might cause the problem known as ―the curse of dimensionality‖. Consequently, to ease this challenge this paper aims to achieve high accuracy system with a minimum set of features. The proposed model uses two methods, namely ―SVM features selection‖ and the common ―Correlation-based Feature Subset Selection (CFS)‖ for the feature dimensions reduction part. In addition, two different classifiers, one Support Vector Machine and the other Neural Network are separately adopted to identify the six emotional states of anger, disgust, fear, happiness, sadness and neutral. The method has been verified using Persian (Persian ESD) and German (EMO-DB) emotional speech databases, which yield high recognition rates in both databases. The results show that SVM feature selection method provides better emotional speech-recognition performance compared to CFS and baseline feature set. Moreover, the new system is able to achieve a recognition rate of (99.44%) on the Persian ESD and (87.21%) on Berlin Emotion Database for speaker-dependent classification. Besides, promising result (76.12%) is obtained for speaker-independent classification case; which is among the best-known accuracies reported on the mentioned database relative to its little number of features.

Real-Time Differential Global Poisoning System Stability and Accuracy Improvement by Utilizing Support Vector Machine

Due to errors, accuracy of Global Positioning System is not so high. Therefore, the Real Time Differential Global Poisoning System (RTDGPS) which is based on the successive transmission message of RTCM protocol, is using in real-time applications. Stability and accuracy of the system, significantly depends to a fast transmission of correction messages. These messages come from the reference station to the user stations and affected by the errors related to each satellite. Receiving correction factors which are transmitted by the reference station are facing with time-lag problem, which can increase the error of the RTDGPS. To overcome this problem, prediction algorithms are used. In this research, support vector machine (SVM) model is used to predict the pseudo range correction. Unfortunately, the practical use of SVM is limited because the quality of SVM models depends on a proper setting of SVM and SVM kernel parameters. Therefore, to determine the main parameters of the SVM, both particle swarm optimization (PSO) and genetic algorithm (GA), as two optimization techniques, are used. The proposed methodology has been implemented by a 6-s predicts time step. Simulations showed that the accuracy of GA–SVM and PSO–SVM are equal to 0.186 and 0.154, respectively.

Differentially Expressed and Prognostically Significant Lincrnas May Impact Immune System and Tumor Progression in Multiple Myeloma (MM)

Long intergenic non-coding RNA (lincRNA) are transcripts longer than 200 nucleotides which have a diverse sets of regulatory functions but do not get translated into protein. lincRNAs are located between the protein coding genes and do not overlap exons of either protein-coding or other non-lincRNA. However precise role of individual lincRNA in disease biology remains unclear. Here, we have evaluated the lincRNA expression and their potential biological functions in MM.We performed RNA-seq on CD138+ MM cells from 296 newly diagnosed patients and 16 normal bone marrow plasma cells (NBM) and analyzed for lincRNA expression. Data from paired-end RNAseq reads were mapped to the latest human genome, differentially expressed lincRNAs were identified and for each expressed lincRNA event free survival was examined with univariate cox regression model and support vector machine. Finally, we identified protein coding genes that are strongly correlated (cor > 0.5) with lincRNAs with significant altered expression in MM and impact on EFS to identify their biological role.lincRNA and protein coding genes that have more than 10 reads/million reads for at least 15 normal samples or 62 MM samples (20% all MM samples) were included in the analysis. We identified 60 differentially expressed lincRNA (adj p value <0.05), 51 of those had at least 1.5 fold change difference. The differentially expressed lncRNAs were in close proximity of Ig-related genes, genome stability related genes, hosting miRNAs such as mir222 and mir22 and previously reported for other cancers (PVT and TTY15). We evaluated relation of these lincRNAs with event free survival (EFS) and observed 6 lincRNAs associated with shorter EFS. We have developed multivariate signature model to predict EFS by using these 6 lincRNAs. We divided our dataset into training (n=99) and test (n=156) dataset and we utilized support vector machine classification to divide samples into 2 groups using six lincRNAs. This model was able to predict good and poor survival groups in training dataset (p val < 0.001) as well as test dataset (p val = 0.002) (Figure).We examined genome wide correlation between these six differentially expressed and prognostically significant lincRNAs to expressed protein coding genes to identify their biological functions in MM. Four of these lincRNAs strongly correlated with 47 to 504 genes (abs(cor) > 0.5), affecting immune system pathways and pathways in cancer including Jak-STAT signaling pathway. We also found that these lincRNAs are also highly correlated with tumor development genes such as TNFRSF1B,FGR,TP53BP2,TNF and T or B cells related genes PIK3CD, BCL6. In addition, two of these lincRNAs (LINC00936 and CTB-61M7.2) were found highly correlated with their protein coding neighbor genes ATP2B1(cor = 0.45) and FCAR (cor = 0.95) respectively and MIR22HG was host gene for mir22 which may indicate lincRNAs are using different machinery in MM to regulate protein coding genes.In summary, we report that lincRNA is differentially expressed and prognostically significant in myeloma and may function through their impact on immune system and tumor progression. Our ongoing integrative approach will provide further evidence of their regulatory role in MM with potential therapeutic application. [Display omitted] DisclosuresAnderson:acetylon pharmaceuticals: Equity Ownership; Celgene Corporation: Consultancy; Gilead: Consultancy; Oncocorp: Equity Ownership; Millennium: Consultancy; BMS: Consultancy. Munshi:onyx: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees; novartis: Membership on an entity's Board of Directors or advisory committees; millenium: Membership on an entity's Board of Directors or advisory committees.

Analysis of nervous fiber, muscle, and blood vessels using their ulraviolet near infrared reflectance characteristics.

Injury to the nervous system can lead to irreversible problems as nervous tissues have limited regenerative capability. Therefore it is imperative to find an objective, reliable, cheap, and easy-to-apply method that separates nervous fibers from muscles and blood vessels. The aim of this study is to determine structural differences that can aid in easy and reliable identification of nervous fibers. We analyzed light reflectance from these tissues from 230 nm to 1000 nm and found that in the range of 400 nm-600 nm nervous fibers have higher reflectance in comparison to others. Therefore, we generated distinct features in this range and utilized support vector machine to automatically classify samples. Classification performance demonstrated that light reflectance is a good candidate feature that can help to classify nervous tissue.

Development of sleep apnea syndrome screening algorithm by using heart rate variability analysis and support vector machine.

Although sleep apnea syndrome (SAS) is a common sleep disorder, most patients with sleep apnea are undiagnosed and untreated because it is difficult for patients themselves to notice SAS in daily living. Polysomnography (PSG) is a gold standard test for sleep disorder diagnosis, however PSG cannot be performed in many hospitals. This fact motivates us to develop an SAS screening system that can be used easily at home. The autonomic nervous function of a patient changes during apnea. Since changes in the autonomic nervous function affect fluctuation of the R-R interval (RRI) of an electrocardiogram (ECG), called heart rate variability (HRV), SAS can be detected through monitoring HRV. The present work proposes a new HRV-based SAS screening algorithm by utilizing support vector machine (SVM), which is a well-known pattern recognition method. In the proposed algorithm, various HRV features are derived from RRI data in both apnea and normal respiration periods of patients and healthy people, and an apnea/normal respiration (A/N) discriminant model is built from the derived HRV features by SVM. The result of applying the proposed SAS screening algorithm to clinical data demonstrates that it can discriminate patients with sleep apnea and healthy people appropriately. The sensitivity and the specificity of the proposed algorithm were 100% and 86%, respectively.

Unlocking Smart Phone through Handwaving Biometrics

Screen locking/unlocking is important for modern smart phones to avoid the unintentional operations and secure the personal stuff. Once the phone is locked, the user should take a specific action or provide some secret information to unlock the phone. The existing unlocking approaches can be categorized into four groups: motion, password, pattern, and fingerprint. Existing approaches do not support smart phones well due to the deficiency of security, high cost, and poor usability. We collect 200 users' handwaving actions with their smart phones and discover an appealing observation: the waving pattern of a person is kind of unique, stable and distinguishable. In this paper, we propose OpenSesame, which employs the users' waving patterns for locking/unlocking. The key feature of our system lies in using four fine-grained and statistic features of handwaving to verify users. Moreover, we utilize support vector machine (SVM) for accurate and fast classification. Our technique is robust compatible across different brands of smart phones, without the need of any specialized hardware. Results from comprehensive experiments show that the mean false positive rate of OpenSesame is around 15 percent, while the false negative rate is lower than 8 percent.

Selective paired ion contrast analysis: a novel algorithm for analyzing postprocessed LC-MS metabolomics data possessing high experimental noise.

One of the consequences in analyzing biological data from noisy sources, such as human subjects, is the sheer variability of experimentally irrelevant factors that cannot be controlled for. This holds true especially in metabolomics, the global study of small molecules in a particular system. While metabolomics can offer deep quantitative insight into the metabolome via easy-to-acquire biofluid samples such as urine and blood, the aforementioned confounding factors can easily overwhelm attempts to extract relevant information. This can mar potentially crucial applications such as biomarker discovery. As such, a new algorithm, called Selective Paired Ion Contrast (SPICA), has been developed with the intent of extracting potentially biologically relevant information from the noisiest of metabolomic data sets. The basic idea of SPICA is built upon redefining the fundamental unit of statistical analysis. Whereas the vast majority of algorithms analyze metabolomics data on a single-ion basis, SPICA relies on analyzing ion-pairs. A standard metabolomic data set is reinterpreted by exhaustively considering all possible ion-pair combinations. Statistical comparisons between sample groups are made only by analyzing the differences in these pairs, which may be crucial in situations where no single metabolite can be used for normalization. With SPICA, human urine data sets from patients undergoing total body irradiation (TBI) and from a colorectal cancer (CRC) relapse study were analyzed in a statistically rigorous manner not possible with conventional methods. In the TBI study, 3530 statistically significant ion-pairs were identified, from which numerous putative radiation specific metabolite-pair biomarkers that mapped to potentially perturbed metabolic pathways were elucidated. In the CRC study, SPICA identified 6461 statistically significant ion-pairs, several of which putatively mapped to folic acid biosynthesis, a key pathway in colorectal cancer. Utilizing support vector machines (SVMs), SPICA was also able to unequivocally outperform binary classifiers built from classical single-ion feature based SVMs.

Support vector machine-based optimized decision threshold adjustment strategy for classifying imbalanced data

Class imbalance problem occurs when the number of training instances belonging to different classes are clearly different. In this scenario, many traditional classifiers often fail to provide excellent enough classification performance, i.e., the accuracy of the majority class is usually much higher than that of the minority class. In this article, we consider to deal with class imbalance problem by utilizing support vector machine (SVM) classifier with an optimized decision threshold adjustment strategy (SVM-OTHR), which answers a puzzled question: how far the classification hyperplane should be moved towards the majority class? Specifically, the proposed strategy is self-adapting and can find the optimal moving distance of the classification hyperplane according to the real distributions of training samples. Furthermore, we also extend the strategy to develop an ensemble version (EnSVM-OTHR) that can further improve the classification performance. Two proposed algorithms are both compared with many state-of-the-art classifiers on 30 skewed data sets acquired from Keel data set Repository by using two popular class imbalance evaluation metrics: F-measure and G-mean. The statistical results of the experiments indicate their superiority.

Abstract 2830: Predictive biomarker identification for response to vantictumab (OMP-18R5; anti-Frizzled) by mining gene expression data of human breast cancer xenografts

Abstract The Wnt/β-catenin signaling pathway has been shown to play key roles in both normal development and tumorigenesis (Polakis, 2007; MacDonald et al., 2009). We have developed a monoclonal antibody, vantictumab, that blocks canonical WNT/β-catenin signaling through binding of five Fzd receptors (1, 2, 5, 7, 8) at a conserved epitope within the extracellular domain. This antibody inhibits the growth of several tumor types, including breast, pancreas, colon and lung. Furthermore, studies also showed that vantictumab reduces tumor-initiating cell frequency and exhibits synergistic activity with standard-of-care chemotherapeutic agents (Gurney et al., 2012). Predictive biomarkers are central to maximizing clinical benefit by targeting breast cancer patients most likely to respond to vantictumab. We analyzed microarray gene expression data from 8 minimally passaged breast cancer xenograft models (mostly triple-negative) with established in vivo responses to vantictumab combined with paclitaxel (4 responders, 4 non-responders). We utilized support vector machine—recursive feature elimination (SVM-RFE, Guyon et al., 2002) to identify genes that can distinguish between responder and non-responders and SVM for classification. Leave-one-out cross-validation was used to measure positive predictive value (PPV), negative predictive value (NPV), sensitivity and specificity of the models. The selected 6-gene signature achieved the best performance with PPV=NPV=sensitivity=specificity=100% in the 8 breast cancer models. In addition, we observed a strong correlation between the gene signature biomarker and the ratio of tumor volume (RTV) observed in the breast xenograft experiments. The identified 6-gene biomarker was used to predict the response to vantictumab in combination with paclitaxel in 6 additional, HER2-negative breast cancer xenograft models. The efficacy in all 6 models was predicted successfully by the biomarker. Prevalence data for the biomarker will be presented for both HER2-negative and triple-negative breast cancers. The 6-gene biomarker is currently being evaluated in a Phase 1b study of vantictumab in combination with paclitaxel in patients with locally recurrent or metastatic HER2-negative breast cancer. Citation Format: Chun Zhang, Pete Yeung, Lucia Beviglia, Belinda Cancilla, Tracy Tang, Wan-Ching Yen, Austin Gurney, John Lewicki, Timothy Hoey, Ann M. Kapoun. Predictive biomarker identification for response to vantictumab (OMP-18R5; anti-Frizzled) by mining gene expression data of human breast cancer xenografts. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2830. doi:10.1158/1538-7445.AM2014-2830

Physiology-based diagnosis algorithm for arteriovenous fistula stenosis detection.

In this paper, a diagnosis algorithm for arteriovenous fistula (AVF) stenosis is developed based on auscultatory features, signal processing, and machine learning. The AVF sound signals are recorded by electronic stethoscopes at pre-defined positions before and after percutaneous transluminal angioplasty (PTA) treatment. Several new signal features of stenosis are identified and quantified, and the physiological explanations for these features are provided. Utilizing support vector machine method, an average of 90% two-fold cross-validation hit-rate can be obtained, with angiography as the gold standard. This offers a non-invasive easy-to-use diagnostic method for medical staff or even patients themselves for early detection of AVF stenosis.

A new feature extraction framework based on wavelets for breast cancer diagnosis

This paper investigates a pattern recognition framework in order to determine and classify breast cancer cases. Initially, a two-class separation study classifying normal and abnormal (cancerous) breast tissues is achieved. The Histogram of Oriented Gradients (HOG), Dense Scale Invariant Feature Transform (DSIFT), and Local Configuration Pattern (LCP) methods are used to extract the rotation- and scale-invariant features for all tissue patches. A classification is made utilizing Support Vector Machine (SVM), k-Nearest Neighborhood (k-NN), Decision Tree, and Fisher Linear Discriminant Analysis (FLDA) via 10-fold cross validation. Then, a three-class study (normal, benign, and malignant cancerous cases) is carried out using similar procedures in a two-class case; however, the attained classification accuracies are not sufficiently satisfied. Therefore, a new feature extraction framework is proposed. The feature vectors are again extracted with this new framework, and more satisfactory results are obtained. Our new framework achieved a remarkable increase in recognition performance for the three-class study.

A systems pharmacology approach to model tyrosine kinase inhibitor‐induced cardiotoxicity gene interaction networks (844.17)

This study implements an integrative bioinformatics workflow to predict gene networks underlying TKI‐induced cardiac adverse events (AEs). The Ontology of Adverse Events (OAE) assisted Natural Language Processing (NLP) algorithm analyzes PubMed abstracts for gene interactions based on the gene centrality in a network.The algorithm utilizes support vector machine (SVM) learning to determine gene pairs of interaction. The gene interactions were mined in association with TKI‐induced cardiac AEs identified by disproportionality analysis of the FDA Adverse Event Reporting System data. These TKIs include dasatinib, imatinib, lapatinib, cetuximab, and trastuzumab. The result gene interaction networks were classified and interpreted for their relevance to the established knowledge confirmed by published experiments. Significant findings include mediator genes linking JUN and ABL1 to JAK‐STAT pathway, and inflammatory response mechanism.

Research on virus detection technique based on ensemble neural network and SVM

Computer viruses have become a serious threat to the information system. Due to the complexity and behavioral uncertainty of virus codes, as well as the emergence of encryption and metamorphic viruses which lead to the ineffectiveness of traditional detection methods, applying artificial intelligence based approach to virus detection has become the focal issue of the current antivirus research. In this paper, we propose a novel approach that introduces ensemble learning into automatic virus detection technique, which is based on the integration of dynamic virus detection and static detection. The detection system utilizes support vector machine (SVM) as member classifier to construct the dynamic behavior model of viruses, and also uses probabilistic neural network (NN) as member classifier for static behavior modeling. Finally, the detection results from all member classifiers are integrated by D–S theory of evidence. The experiments show that the diversity of combining heterogeneous classifiers leads to the great performance improvement of the ensemble method of virus detector. The experimental results show that the proposed approach is very efficient in detecting unknown and metamorphic viruses, and further comparison indicates that its performance is superior to most of the popular commercial antivirus tools.

Predicting Active Users' Personality Based on Micro-Blogging Behaviors

Because of its richness and availability, micro-blogging has become an ideal platform for conducting psychological research. In this paper, we proposed to predict active users' personality traits through micro-blogging behaviors. 547 Chinese active users of micro-blogging participated in this study. Their personality traits were measured by the Big Five Inventory, and digital records of micro-blogging behaviors were collected via web crawlers. After extracting 845 micro-blogging behavioral features, we first trained classification models utilizing Support Vector Machine (SVM), differentiating participants with high and low scores on each dimension of the Big Five Inventory. The classification accuracy ranged from 84% to 92%. We also built regression models utilizing PaceRegression methods, predicting participants' scores on each dimension of the Big Five Inventory. The Pearson correlation coefficients between predicted scores and actual scores ranged from 0.48 to 0.54. Results indicated that active users' personality traits could be predicted by micro-blogging behaviors.