Linear Support Vector Classifier Research Articles

Sentiment analysis of extremism in social media from textual information

Uncertainty in political, religious, and social issues causes extremism among people that are depicted by their sentiments on social media. Although, English is the most common language used to share views on social media, however, other vicinity based languages are also used by locals. Thus, it is also required to incorporate the views in such languages along with widely used languages for revealing better insights from data. This research focuses on the sentimental analysis of social media multilingual textual data to discover the intensity of the sentiments of extremism. Our study classifies the incorporated textual views into any of four categories, including high extreme, low extreme, moderate, and neutral, based on their level of extremism. Initially, a multilingual lexicon with the intensity weights is created. This lexicon is validated from domain experts and it attains 88% accuracy for validation. Subsequently, Multinomial Naïve Bayes and Linear Support Vector Classifier algorithms are employed for classification purposes. Overall, on the underlying multilingual dataset, Linear Support Vector Classifier out-performs with an accuracy of 82%.

ОЦІНКА ЕФЕКТИВНОСТІ МЕТОДІВ САМООЦІНЮВАННЯ ЗОБРАЖЕННЯ

Lighting Normalization is an especially important issue in the image recognitions systems since different illumination conditions can significantly change the recognition results, and the lighting normalization allows minimizing negative effects of various illumination conditions. In this paper, we are evaluating the recognition performance of several lighting normalization methods based on the Self-Quotient ImagE(SQI) method introduced by Haitao Wang, Stan Z. Li, Yangsheng Wang, and Jianjun Zhang. For evaluation, we chose the original implementation and the most perspective latest modifications of the original SQI method, including the Gabor Quotient ImagE(GQI) method introduced by Sanun Srisuk and Amnart Petpon in 2008, and the Fast Self-Quotient ImagE(FSQI) method and its modifications proposed by authors in previous works. We are proposing an evaluation framework which uses the Cropped Extended Yale Face Database B, which allows showing the difference of the recognition results for different illumination conditions. Also, we are testing all results using two classifiers: Nearest Neighbor Classifier and Linear Support Vector Classifier. This approach allows us not only to calculate recognition accuracy for each method and select the best method but also show the importance of the proper choice of the classification method, which can have a significant influence on recognition results. We were able to show the significant decreasing of recognition accuracy for un-processed (RAW) images with increasing the angle between the lighting source and the normal to the object. From the other side, our experiments had shown the almost uniform distribution of the recognition accuracy for images processed by lighting normalization methods based on the SQI method. Another showed but expected result represented in this paper is the increasing of the recognition accuracy with the increasing of the filter kernel size. However, the large filter kernel sizes are much more computationally expensive and can produce negative effects on output images. Also, we were shown in our experiments, that the second modification of the FSQI method, called FSQI3, is better almost in all cases for all filter kernel sizes, especially, if we use Linear Support Vector Classifier for classification.

Abnormal Event Detection via Feature Expectation Subgraph Calibrating Classification in Video Surveillance Scenes

At present, the existing abnormal event detection models based on deep learning mainly focus on data represented by a vectorial form, which pay little attention to the impact of the internal structure characteristics of feature vector. In addition, a single classifier is difficult to ensure the accuracy of classification. In order to address the above issues, we propose an abnormal event detection hybrid modulation method via feature expectation subgraph calibrating classification in video surveillance scenes in this paper. Our main contribution is to calibrate the classification of a single classifier by constructing feature expectation subgraphs. First, we employ convolutional neural network and long short-term memory models to extract the spatiotemporal features of video frame, and then construct the feature expectation subgraph for each key frame of every video, which could be used to capture the internal sequential and topological relational characteristics of structured feature vector. Second, we project expectation subgraphs on the sparse vector to combine with a support vector classifier to calibrate the results of a linear support vector classifier. Finally, the experiments on a common dataset named UCSDped1 and a coal mining video dataset in comparison with some existing works demonstrate that the performance of the proposed method is better than several the state-of-the-art approaches.

ACTSMLT: Automatic Classification of Text Summarization using Machine Learning Technique

In today’s world, due to the steep rise in internet users, Community Question Answering (CQA) has attracted many research communities. In order to provide the correct and perfect answer to the user asked question from a given large collection of text data, understanding the question properly to suggest a precise answer is a challenging task. Therefore, Question Answering (QA) system is a challenging task than a common information retrieval task done by many search engines. In this paper, an automatic prediction of the quality of CQA answers is proposed. This is accomplished by using five well known machine learning algorithms. Usually, questions asked by the user are based on a topic or theme. We try to exploit this feature in our work by identifying the category of the question posted and further map with the corresponding question. Similarly, for the answers posted by the multiple user’s are processed as answer for category mapping. Here, the results show that for Question Classification (QA), Linear Support Vector Classification (LSVC) is found to be the best classifier and Multinomial Logistic Regression (MLR) is the most suitable for Answer Classification (AC). The MS Macro dataset is used as the underlying dataset for retrieving and testing the question and answer classifiers. The Yahoo Answers are used as a golden reference during the testing throughout our experiments. Experiments results show that the proposed technique is efficient and outperforms Metzler and Kanungo’s (MK++) [1] while providing the best answer summary satisfying the user’s queries.

Predicting the occurrence of surgical site infections using text mining and machine learning.

In this study we propose the use of text mining and machine learning methods to predict and detect Surgical Site Infections (SSIs) using textual descriptions of surgeries and post-operative patients’ records, mined from the database of a high complexity University hospital. SSIs are among the most common adverse events experienced by hospitalized patients; preventing such events is fundamental to ensure patients’ safety. Knowledge on SSI occurrence rates may also be useful in preventing future episodes. We analyzed 15,479 surgery descriptions and post-operative records testing different preprocessing strategies and the following machine learning algorithms: Linear SVC, Logistic Regression, Multinomial Naive Bayes, Nearest Centroid, Random Forest, Stochastic Gradient Descent, and Support Vector Classification (SVC). For prediction purposes, the best result was obtained using the Stochastic Gradient Descent method (79.7% ROC-AUC); for detection, Logistic Regression yielded the best performance (80.6% ROC-AUC).

Automating fake news detection system using multi-level voting model

The issues of online fake news have attained an increasing eminence in the diffusion of shaping news stories online. Misleading or unreliable information in the form of videos, posts, articles, URLs is extensively disseminated through popular social media platforms such as Facebook and Twitter. As a result, editors and journalists are in need of new tools that can help them to pace up the verification process for the content that has been originated from social media. Motivated by the need for automated detection of fake news, the goal is to find out which classification model identifies phony features accurately using three feature extraction techniques, Term Frequency–Inverse Document Frequency (TF–IDF), Count-Vectorizer (CV) and Hashing-Vectorizer (HV). Also, in this paper, a novel multi-level voting ensemble model is proposed. The proposed system has been tested on three datasets using twelve classifiers. These ML classifiers are combined based on their false prediction ratio. It has been observed that the Passive Aggressive, Logistic Regression and Linear Support Vector Classifier (LinearSVC) individually perform best using TF-IDF, CV and HV feature extraction approaches, respectively, based on their performance metrics, whereas the proposed model outperforms the Passive Aggressive model by 0.8%, Logistic Regression model by 1.3%, LinearSVC model by 0.4% using TF-IDF, CV and HV, respectively. The proposed system can also be used to predict the fake content (textual form) from online social media websites.

Effective K-nearest neighbor classifications for Wisconsin breast cancer data sets

ABSTRACTUsing machine learning algorithms for early prediction of the signs and symptoms of breast cancer is in demand nowadays. One of these algorithms is the K-nearest neighbor (KNN), which uses a technique for measuring the distance among data. The performance of KNN depends on the number of neighboring elements known as the K value. This study involves the exploration of KNN performance by using various distance functions and K values to find an effective KNN. Wisconsin breast cancer (WBC) and Wisconsin diagnostic breast cancer (WDBC) datasets from the UC Irvine machine learning repository were used as our main data sources. Experiments with each dataset were composed of three iterations. The first iteration of the experiment was without feature selection. The second one was the L1-norm based selection from the model, which used the linear support vector classifier feature selection, and the third iteration was with Chi-square-based feature selection. Numerous evaluation metrics like accuracy, receiver operating characteristic (ROC) curve with the area under curve (AUC) and sensitivity, etc., were used for the assessment of the implemented techniques. The results indicated that the technique involving the Chi-square-based feature selection achieved the highest accuracy with the Canberra or Manhattan distance functions for both datasets. The optimal K values for these distance functions ranged from 1 to 9. This study indicated that with the appropriate selection of the K value and a distance function in KNN, the Chi-square-based feature selection for the WBC datasets gives the highest accuracy rate as compared with the existing models.Abbreviations: KNN: K-nearest neighbor; Chi2: Chi-square; WBC: Wisconsin breast cancer

Radiomics features on non-contrast computed tomography predict early enlargement of spontaneous intracerebral hemorrhage

ObjectiveTo explore the value of radiomics features on non-contrast computed tomography (NCCT) in predicting early enlargement of spontaneous intracerebral hemorrhage (SICH). Patients and Methods167 patients with SICH were divided into enlarged hematoma and non-enlarged hematoma groups based on the volume of hematoma on 24-h follow-up CT images > 30% and/or 6 ml of the baseline NCCT. The baseline NCCT images of all cases were imported into radiomics software to extract the radiomics features of the initial hematoma. For each case, the features with good predictability were retained after the feature-selected process; the remaining features were used to construct model with 23 algorithms one-by-one. A 5-fold method was used to cross-validate the model and repeated 5 times. The algorithm model with the highest accuracy was selected as predictive model for hematoma enlargement (HE) in SICH, its average parameters including AUC, accuracy, sensitivity, specificity, F1 score, positive predictive value (PPV), negative predictive value (NPV), false positive rate (FPR), false negative rate (FNR),and false discovery rate (FDR) were taken as evaluating indicators. ResultsA total of 1227 texture features of each cerebral hematoma were obtained. After the feature-selected process, 4 features (wavelet-LHL mean, wavelet-LLL _ Idm, wavelet-LLL _run length non-uniformity normalized, and wavelet-LLL _contrast) remained to construct the predictive models. Among 23 model algorithms, Linear Support Vector Classifier showed the highest accuracy (72.6%), and eventually was selected as the predictive model, its AUC, accuracy, sensitivity, specificity, F1 score, PPV, NPV, FPR, FNR, and FDR were 0.729, 0.726,0.717,0.736,0.714, 0.736, 0.741, 0.264, 0.283 and 0.264, respectively. ConclusionRadiomics features of cerebral hematoma on baseline NCCT images showed good performance in predicting HE of SICH.

Predicting VMAT patient-specific QA results using a support vector classifier trained on treatment plan characteristics and linac QC metrics

The use of treatment plan characteristics to predict patient-specific quality assurance (QA) measurement results has recently been reported as a strategy to help facilitate automated pre-treatment verification workflows or to provide a virtual assessment of delivery quality. The goal of this work is to investigate the potential of using treatment plan characteristics and linac performance metrics (i.e. quality control test results) in combination with machine learning techniques to predict the results of VMAT patient-specific QA measurements.Using features that describe treatment plan complexity and linac performance metrics, we trained a linear support vector classifier (SVC) to classify the results of VMAT patient-specific QA measurements. The ‘targets’ in this model were simple classes representing median dose difference between measured and expected dose distributions—‘hot’ if the median dose deviation was >1%, ‘cold’ if it was <−1%, and ‘normal’ if it was within ±1%. A total of 1620 unique patient-specific QA measurements were available for model development and testing. 75% of the data were used to develop and cross-validate the model, and the remaining 25% were used for an independent assessment of model performance. For the model development phase, a recursive feature elimination (RFE) cross-validation technique was used to eliminate unimportant features. Model performance was assessed using receiver operator characteristic (ROC) curve metrics.Of the ten features found to be most predictive of patient-specific QA measurement results, half were derived from treatment plan characteristics and half from quality control (QC) metrics characterizing linac performance. The model achieved a micro-averaged area under the ROC curve of 0.93, and a macro-averaged area under the ROC curve of 0.88.This work demonstrates the potential of using both treatment plan characteristics and routine linac QC results in the development of machine learning models for VMAT patient-specific QA measurements.

A Temporal Neural Trace of Wavelet Coefficients in Human Object Vision: An MEG Study.

Wavelet transform has been widely used in image and signal processing applications such as denoising and compression. In this study, we explore the relation of the wavelet representation of stimuli with MEG signals acquired from a human object recognition experiment. To investigate the signature of wavelet descriptors in the visual system, we apply five levels of multi-resolution wavelet decomposition to the stimuli presented to participants during MEG recording and extract the approximation and detail sub-bands (horizontal, vertical, diagonal) coefficients in each level of decomposition. Apart from, employing multivariate pattern analysis (MVPA), a linear support vector classifier (SVM) is trained and tested over the time on MEG pattern vectors to decode neural information. Then, we calculate the representational dissimilarity matrix (RDM) on each time point of the MEG data and also on wavelet descriptors using classifier accuracy and one minus Pearson correlation coefficient, respectively. Given the time-courses calculated from performing the Pearson correlation between the wavelet descriptors RDMs and MEG decoding accuracy in each time point, our result shows that the peak latency of the wavelet approximation time courses occurs later for higher level coefficients. Furthermore, studying the neural trace of detail sub-bands indicates that the overall number of statistically significant time points for the horizontal and vertical detail coefficients is noticeably higher than diagonal detail coefficients, confirming the evidence of the oblique effect that the horizontal and vertical lines are more decodable in the human brain.

Sensorless Synchronous Motors Classification Using Random Forest and Linear Support Vector Classifier

Automation system become increasingly complex and it needed to be monitored actively. Due to wide variance tasks to be done not suprising that the complexity problem occurs. The failure of the system can impact to economic loss for a company. For this purpose, this paper compare the methods to classify sensorless synchronous motors task to investigate the accuracy and precision of random forest classification and linear Support Vector Classifier (SVC). With cross-validation, the result show that the average accuracy is 99.816% with random forest and 66.04% with linear SVC. As the conclution, random forest classification is more precise than SCV method in this field.

Exploring Deep Spectrum Representations via Attention-Based Recurrent and Convolutional Neural Networks for Speech Emotion Recognition

The automatic detection of an emotional state from human speech, which plays a crucial role in the area of human–machine interaction, has consistently been shown to be a difficult task for machine learning algorithms. Previous work on emotion recognition has mostly focused on the extraction of carefully hand-crafted and highly engineered features. Results from these works have demonstrated the importance of discriminative spatio-temporal features to model the continual evolutions of different emotions. Recently, spectrogram representations of emotional speech have achieved competitive performance for automatic speech emotion recognition (SER). How machine learning algorithms learn the effective compositional spatio-temporal dynamics for SER has been a fundamental problem of deep representations, herein denoted as deep spectrum representations. In this paper, we develop a model to alleviate this limitation by leveraging a parallel combination of attention-based bidirectional long short-term memory recurrent neural networks with attention-based fully convolutional networks (FCN). The extensive experiments were undertaken on the interactive emotional dyadic motion capture (IEMOCAP) and FAU aibo emotion corpus (FAU-AEC) to highlight the effectiveness of our approach. The experimental results indicate that deep spectrum representations extracted from the proposed model are well-suited to the task of SER, achieving a WA of 68.1% and a UA of 67.0% on IEMOCAP, and 45.4% for UA on FAU-AEC dataset. Key results indicate that the extracted deep representations combined with a linear support vector classifier are comparable in performance with eGeMAPS and COMPARE, two standard acoustic feature representations.

Comparison of Feature Selection Methods and Machine Learning Classifiers for Radiomics Analysis in Glioma Grading

Radiomics-based researches have shown predictive abilities with machine-learning approaches. However, it is still unknown whether different radiomics strategies affect the prediction performance. The aim of this study was to compare the prediction performance of frequently utilized radiomics feature selection and classification methods in glioma grading. Quantitative radiomics features were extracted from tumor regions in 210 Glioblastoma (GBM) and 75 low-grade glioma (LGG) MRI subjects. Then, the diagnostic performance of sixteen feature selection and fifteen classification methods were evaluated by using two different test modes: ten-fold cross-validation and percentage split. Balanced accuracy and area under the curve (AUC) of the receiver operating characteristic were used to evaluate prediction performance. In addition, the roles of the number of selected features, feature type, MRI modality, and tumor sub-region were compared to optimize the radiomics-based prediction. The results indicated that the combination of feature selection method L 1 -based linear support vector machine (L 1 -SVM) and classifier multi-layer perceptron (MLPC) achieved the best performance in the differentiation of GBM and LGG in both ten-fold cross validation (balanced accuracy:0.944, AUC:0.986) and percentage split (balanced accuracy:0.953, AUC:0.981). For radiomics feature extraction, the enhancing tumor region (ET) combined with necrotic and non-enhancing tumor (NCR/NET) regions in T1 post-contrast (T1-Gd) modality provided more considerable tumor-related phenotypes than other combinations of tumor region and MRI modality. Our comparative investigation indicated that both feature selection methods and machine learning classifiers affected the predictive performance in glioma grading. Also, the cross-combination strategy for comparison of radiomics feature selection and classification methods provided a way of searching optimal machine learning model for future radiomics-based prediction.

Windows malware detection system based on LSVC recommended hybrid features

To combat exponentially evolved modern malware, an effective Malware Detection System and precise malware classification is highly essential. In this paper, the Linear Support Vector Classification (LSVC) recommended Hybrid Features based Malware Detection System (HF-MDS) has been proposed. It uses a combination of the static and dynamic features of the Portable Executable (PE) files as hybrid features to identify unknown malware. The application program interface calls invoked by the PE files during their execution along with their correspondent category are collected and considered as dynamic features from the PE file behavioural report produced by the Cuckoo Sandbox. The PE files’ header details such as optional header, disk operating system header, and file header are treated as static features. The LSVC is used as a feature selector to choose prominent static and dynamic features from their respective Original Feature Space. The features recommended by the LSVC are highly discriminative and used as final features for the classification process. Different sets of experiments were conducted using real-world malware samples to verify the combination of static and dynamic features, which encourage the classifier to attain high accuracy. The tenfold cross-validation experimental results demonstrate that the proposed HF-MDS is proficient in precisely detecting malware and benign PE files by attaining detection accuracy of 99.743% with sequential minimal optimization classifier consisting of hybrid features.

T18. MULTIVARIATE BRAIN ANATOMICAL DIFFERENCES IN POSITIVE AND NEGATIVE SCHIZOTYPY: PRELIMINARY RESULTS FROM THE TYPIA STUDY

BackgroundIn schizotypy, a factor structure similar to the one observed in schizophrenia has been unraveled, being the positive and negative the most consistently replicated dimensions. Despite this fact, most of the studies on brain volume patterns in schizotypy consider it as an unitary rather than a multidimensional construct. Hence, based on previous results showing that schizophrenia and schizotypal personality traits share common neurodevelopmental patterns, it is hypothesized that brain volumetric patterns in individuals with high positive schizotypy are intrinsically different to those observed in persons reporting high negative schizotypy and to individuals with overall low schizotypal traits. The present study aims to evaluate this hypothesis using novel machine learning techniques to address the multivariate nature of psychotic diseases and the brain itself.MethodsData from the TYPIA Study, an ongoing project conducted at the Ludwig-Maximilian University of Munich and the University of Bonn in Germany, was used to investigate whether brain volumetric patterns are distinct in healthy individuals with high positive (HPS) and high negative schizotypy (HNS) when compared to one another (HPS vs HNS) and to individuals with self-reported low schizotypy (LS vs HNS and LS vs HPS). A preliminary analysis on grey matter volumetric patterns from 29 LS (19 f., mean age: 24.6 y.), 28 HNS (20 f., mean age: 26.8 y.) and 23 HPS (17 f., mean age: 26.4 years) individuals from the general population without any current psychiatric diagnosis was performed. Group divisions were based on the introvertive anhedonia and unusual experiences subscales from the Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE). Structural images were preprocessed with a standard voxel based morphometry pipeline using the SPM-based CAT12 toolbox in Matlab. After age, sex and grey matter intracranial volume and center corrections, a linear support vector classification (SVC) algorithm was used to assess separability between the groups.ResultsOur preliminary cross-validated results showed that LS and HNS can be separated with 56.0 % balanced accuracy (BAC), whereas LS vs HPS and HNS vs HPS allowed for only 42.87% and 48.8% BAC respectively. Interestingly, a post-hoc analysis comparing LS vs both high schizotypy groups merged together showed the highest BAC (59.2%). As expected, the brain differences between groups are rather small, since the sample consists fully of healthy controls. However, these results indicate that personality traits related to HNS are linked to more pronounced changes in the brain as compared to HPS. Nevertheless, schizotypy as a combination of the positive and negative dimensions allowed for a higher classification accuracy when compared to LS, supporting the notion of schizotypy as a unitary construct as observed from the post-hoc analysis. Furthermore, HNS and HPS were not separable by the algorithm, most likely due to the intrinsic heterogeneity of the construct.DiscussionOur results align with previous studies claiming that negative symptoms are associated with structural changes in the CNS whereas positive symptoms relate to changes in functioning and activation of the brain. A larger sample as well as using other data modalities will confirm the stability of our findings. Research on volumetric patterns of the brain areas related to negative symptoms in non-clinical samples might lead to a better understanding of the underlying causes of schizophrenia. Above all, our results show that investigating non-clinical expression of psychosis-like symptoms is a promising strategy to understand the prodromal stadium of schizophrenia.

Resting-state connectivity predicts levodopa-induced dyskinesias in Parkinson's disease.

ABSTRACTBackgroundLevodopa‐induced dyskinesias are a common side effect of dopaminergic therapy in PD, but their neural correlates remain poorly understood.ObjectivesThis study examines whether dyskinesias are associated with abnormal dopaminergic modulation of resting‐state cortico‐striatal connectivity.MethodsTwelve PD patients with peak‐of‐dose dyskinesias and 12 patients without dyskinesias were withdrawn from dopaminergic medication. All patients received a single dose of fast‐acting soluble levodopa and then underwent resting‐state functional magnetic resonance imaging before any dyskinesias emerged. Levodopa‐induced modulation of cortico‐striatal resting‐state connectivity was assessed between the putamen and the following 3 cortical regions of interest: supplementary motor area, primary sensorimotor cortex, and right inferior frontal gyrus. These functional connectivity measures were entered into a linear support vector classifier to predict whether an individual patient would develop dyskinesias after levodopa intake. Linear regression analysis was applied to test which connectivity measures would predict dyskinesia severity.ResultsDopaminergic modulation of resting‐state connectivity between the putamen and primary sensorimotor cortex in the most affected hemisphere predicted whether patients would develop dyskinesias with a specificity of 100% and a sensitivity of 91% (P < .0001). Modulation of resting‐state connectivity between the supplementary motor area and putamen predicted interindividual differences in dyskinesia severity (R 2 = 0.627, P = .004). Resting‐state connectivity between the right inferior frontal gyrus and putamen neither predicted dyskinesia status nor dyskinesia severity.ConclusionsThe results corroborate the notion that altered dopaminergic modulation of cortico‐striatal connectivity plays a key role in the pathophysiology of dyskinesias in PD. © 2016 International Parkinson and Movement Disorder Society

Chemical Discrimination with an Unmodified Graphene Chemical Sensor

A graphene chemical vapor sensor with an unmodified surface has been fabricated and thoroughly characterized upon exposure to headspace vapor of a variety of solvents and related compounds. The vapor sensor exhibits excellent discrimination toward a variety of chemical compounds. Principle component analysis (PCA) was performed to explore the extent of grouping for each compound and separation between compounds and chemical classes. The prediction accuracy of the sensor is evaluated with linear discrimination analysis, k-nearest neighbor, random forest, and support vector classifiers. The combination of PCA and prediction accuracies demonstrates the discrimination capability of an unmodified graphene chemical vapor sensor. Such a vapor sensor is very attractive for application in small, low-power, robust, and adaptable cross-reactive arrays in electronic noses.

Altered regional homogeneity patterns in adults with attention-deficit hyperactivity disorder

PurposeInvestigating the discriminative brain map for patients with attention-deficit/hyperactivity disorder (ADHD) based on feature selection and classifier; and identifying patients with ADHD based on the discriminative model. Materials and methodsA dataset of resting state fMRI contains 23 patients with ADHD and 23 healthy subjects were analyzed. Regional homogeneity (ReHo) was extracted from resting state fMRI signals and used as model inputs. Raw ReHo features were ranked and selected in a loop according to their p values. Selected features were trained and tested by support vector machines (SVM) in a cross validation procedure. Cross validation was repeated in feature selection loop to produce optimized model. ResultsOptimized discriminative map indicated that the ADHD brains exhibit more increased activities than normal controls in bilateral occipital lobes and left front lobe. The altered brain regions included portions of basal ganglia, insula, precuneus, anterior cingulate cortex (ACC), posterior cingulate cortex (PCC), thalamus, and cerebellum. Correlation coefficients indicated significant positive correlation of inattentive scores with bilateral cuneus and precuneus, and significant negative correlation of hyperactive/impulsive scores with bilateral insula and claustrum. Additionally, the optimized model produced total accuracy of 80% and sensitivity of 87%. ConclusionADHD brain regions were more activated than normal controls during resting state. Linear support vector classifier can provide useful discriminative information of altered ReHo patterns for ADHD; and feature selection can improve the performances of classification.

Large-scale linear support vector regression

Support vector regression (SVR) and support vector classification (SVC) are popular learning techniques, but their use with kernels is often time consuming. Recently, linear SVC without kernels has...

PLS-Based Gene Selection and Identification of Tumor-Specific Genes

In view of the characteristics of high-dimensional small sample, strong relevance, and high noise of the identification of tumor-specific genes on microarray, a novel partial least squares (PLS) based gene-selection method, which synthesizes genetic relatedness and is suitable for multicategory classification, is presented. Using the explanation difference of independent variables on dependent variable (class), we define three indicators for global gene selection, which takes into accounts the combined effects of all the genes and the correlation among the genes. Integrated with the linear kernel support vector classifier (SVC), the proposed method is tested by MIT acute myeloid leukemia/acute lymphoblastic leukemia (AML/ALL) and small round blue cell tumors (SRBCT) data sets. A subset of specific genes with small numbers and high identification are obtained. The results indicate that our proposed PLS-based method for tumor-specific genes selection is highly efficient. Compared to the literature, the selected specific genes from both two-category dataset AML/ALL and multicategory dataset SRBCT are credible. Further investigation shows that the proposed gene-selection method is robust. Overall, the proposed method can effectively solve feature-selection problem on high-dimensional small sample. At the same time, it has good performance for multicategory classification as well.