Support Vector Machine Prediction Research Articles

Burst and principal components analyses of MEA data for 16 chemicals describe at least three effects classes

Microelectrode arrays (MEAs) can be used to detect drug and chemical induced changes in neuronal network function and have been used for neurotoxicity screening. As a proof-of-concept, the current study assessed the utility of analytical “fingerprinting” using principal components analysis (PCA) and chemical class prediction using support vector machines (SVMs) to classify chemical effects based on MEA data from 16 chemicals. Spontaneous firing rate in primary cortical cultures was increased by bicuculline (BIC), lindane (LND), RDX and picrotoxin (PTX); not changed by nicotine (NIC), acetaminophen (ACE), and glyphosate (GLY); and decreased by muscimol (MUS), verapamil (VER), fipronil (FIP), fluoxetine (FLU), chlorpyrifos oxon (CPO), domoic acid (DA), deltamethrin (DELT) and dimethyl phthalate (DMP). PCA was performed on mean firing rate, bursting parameters and synchrony data for concentrations above each chemical's EC50 for mean firing rate. The first three principal components accounted for 67.5, 19.7, and 6.9% of the data variability and were used to identify separation between chemical classes visually through spatial proximity. In the PCA, there was clear separation of GABAA antagonists BIC, LND, and RDX from other chemicals. For the SVM prediction model, the experiments were classified into the three chemical classes of increasing, decreasing or no change in activity with a mean accuracy of 83.8% under a radial kernel with 10-fold cross-validation. The separation of different chemical classes through PCA and high prediction accuracy in SVM of a small dataset indicates that MEA data may be useful for separating chemicals into effects classes using these or other related approaches.

Prediction of hot spots residues in protein–protein interface using network feature and microenvironment feature

Hot spots residues in protein–protein interface play crucial roles in protein binding. In the present study, complex network method was applied to uncover influence of neighboring residues on hot spots and then several network and microenvironment features were designed to describe the diversity of environment of hot spots. After feature analysis by permutation importance in Random Forest (RF), an optimal 58-dimensional feature set including ten network and microenvironment features was selected and then applied to construct a Support Vector Machine (SVM) prediction model for hot spots. A satisfactory accuracy (ACC) value of 79.0% and a Mathew's correlation coefficient (MCC) value of 0.470 were obtained for independent test set. The novel network features and microenvironment features were proved to be promising in discovering hot spots in interfaces. A further microenvironment analysis was also performed. Amino acid residues directly contacting with hot spots in residue–residue interaction network exhibit significant importance for the microenvironment of hot spots. Amino acid alanine (A), aspartic acid (D), glycine (G), histidine (H), isoleucine (I), asparagine (N), serine (S) and tyrosine (Y) are more likely to occur in the vicinity of hot spots than in the vicinity of non-hot spots. These amino acid residues probably cluster together to construct a proper microenvironment for hot spots.

Applying Support Vector Machines in Rebound Hammer Test

There are several nondestructive testing techniques available to test the compressive strength of the concrete and the Rebound Hammer Test is among one of the fast and economical methods. Nevertheless, it is found that the prediction results from Rebound Hammer Test are not satisfying (over 20% mean absolute percentage error). In view of this, this research intends to develop a concrete compressive strength prediction model for the SilverSchmidt test hammer, using data collected from 838 lab tests. The Q-values yield from the concrete test hammer SilverSchmidt is set as the input variable and the concrete compressive strength is set as the output variable for the prediction model. For the non-linear relationships, artificial intelligence technique, Support Vector Machines (SVMs), are adopted to develop the prediction models. The results show that the mean absolute percentage errors for SVMs prediction model, 6.76%, improves a lot when comparing to SilverSchmidt predictions. It is recommended that the artificial intelligence prediction models can be applied in the SilverSchmidt tests to improve the prediction accuracy.

Research of Aggregated Wind Power Prediction Method for a Region

Wind power is intermittent and volatility. Some new problems would arise to power system operation when Large-scale wind farm is connected with power systems. One of the most important effect is the influence on the grid dispatch. An aggregated wind power prediction method for a region is presented. By means of analyzing power characteristics and correlation, then the greater correlation is selected as model input. Based on grey correlation theory, a least squares support vector machine prediction model is established. Finally, this method is executed on a real case and integrated wind power prediction method can effectively improve the prediction accuracy and simplify the prediction step are proved.

Predicting Protein Model Quality from Sequence Alignments by Support Vector Machines.

Assessing the quality of a protein structure model is essential for protein structure prediction. Here, we developed a Support Vector Machine (SVM) method to predict the quality score (GDT-TS score) of a protein structure model from the features extracted from the sequence alignment used to generate the model. We developed a Support Vector Machine (SVM) model quality assessment method, taking either a query-single-template pairwise alignment or a query-multitemplate alignment as input. For the pairwise alignment scheme, the input features fed into the SVM predictor include the normalized e-value of the given alignment, the percentage of identical residue pairs in the alignment, the percentage of residues of the query aligned with those of the template, and the sum of the BLOSUM scores of all aligned residues divided by the length of the aligned positions. Similarly, for the multiple-alignment scheme, the input features include the percentage of the residues of the target sequence aligned with those in one or more templates, the percentage of aligned residues of the target sequence that are the same as that of any one template, the average BLOSUM score of aligned residues and the average Gonnet160 score of aligned residues. A SVM regression predictor was trained on the training data to predict the GDT-TS scores of the models from the input features. The Root Mean Square Error (RMSE) and the Absolute Mean Error (ABS) between predicted and real GDT-TS scores were calculated to evaluate the performance. A five-fold cross validation was applied to select the best parameter values based on the average RMSE and ABS on the five folds. The RMSE and ABS of the optimized SVM predictor on the testing data were close to 0.1. The good performance of the SVM and sequence alignment based predictor indicates that integrating sequence alignment features with a SVM is effective for protein model quality assessment.

Optimization of Financial Forecasting Based on Least Squares Support Vector Machine

To master the variation regularity of finance, obtain greater benefits in stock investment. study of the support vector machine and application in prediction of stock market. The simulated annealing algorithm to optimize the least squares support vector machine prediction model, and the least square support vector machine and simulated annealing algorithm is described, given the optimal prediction model. Through the research on the simulation of the Hang Seng Index, shows that this method is simple, fast convergence, the algorithm with high accuracy. Has the actual guiding sense for investors, the stock market of the financial firm to operate.

Fault Prediction Based on Data Fusion

The performance of hydraulic pump will directly affect the normal work of the entire hydraulic system, so it is essential for its condition monitoring and fault prediction. It collects vibration signal and pressure signal for hydraulic pump which degenerates from the normal state to loose slipper state. It uses wavelet packet to decompose energy of frequency area in order to get fault feature vectors and builds support vector machine prediction model for feature vectors, then predicts possible fault by D-S evidence theory. The experiment results show that the method can effectively predict fault for hydraulic pump.

Multi-Step Ahead Prediction for Electromechanical Device Using Multivariate SVM Predictor

Electromechanical device driven by motor plays a crucial role in the operation of modern industry. A reliable time series predictor can be a very useful tool in the field of Condition-Based Maintenance (CBM) to forecast the behavior of motor. This paper presents an approach to predict the operating conditions of motor based on support vector machine (SVM). In order to improve the accuracy of prediction, a multivariate SVM predictor is built using multi-source multi-attribute information for time series prediction. The advantages of the multivariate SVM predictor is verified by empirical results derived from a real system of motor in power plant.

Computer-Aided Diagnosis for Early-Stage Lung Cancer Based on Longitudinal and Balanced Data

BackgroundLung cancer is one of the most common forms of cancer resulting in over a million deaths per year worldwide. Typically, the problem can be approached by developing more discriminative diagnosis methods. In this paper, computer-aided diagnosis was used to facilitate the prediction of characteristics of solitary pulmonary nodules in CT of lungs to diagnose early-stage lung cancer.MethodsThe synthetic minority over-sampling technique (SMOTE) was used to account for raw data in order to balance the original training data set. Curvelet-transformation textural features, together with 3 patient demographic characteristics, and 9 morphological features were used to establish a support vector machine (SVM) prediction model. Longitudinal data as the test data set was used to evaluate the classification performance of predicting early-stage lung cancer.ResultsUsing the SMOTE as a pre-processing procedure, the original training data was balanced with a ratio of malignant to benign cases of 1∶1. Accuracy based on cross-evaluation for the original unbalanced data and balanced data was 80% and 97%, respectively. Based on Curvelet-transformation textural features and other features, the SVM prediction model had good classification performance for early-stage lung cancer, with an area under the curve of the SVMs of 0.949 (P<0.001). Textural feature (standard deviation) showed benign cases had a higher change in the follow-up period than malignant cases.ConclusionsWith textural features extracted from a Curvelet transformation and other parameters, a sensitive support vector machine prediction model can increase the rate of diagnosis for early-stage lung cancer. This scheme can be used as an auxiliary tool to differentiate between benign and malignant early-stage lung cancers in CT images.

Predicting substrates of the human breast cancer resistance protein using a support vector machine method

BackgroundHuman breast cancer resistance protein (BCRP) is an ATP-binding cassette (ABC) efflux transporter that confers multidrug resistance in cancers and also plays an important role in the absorption, distribution and elimination of drugs. Prediction as to if drugs or new molecular entities are BCRP substrates should afford a cost-effective means that can help evaluate the pharmacokinetic properties, efficacy, and safety of these drugs or drug candidates. At present, limited studies have been done to develop in silico prediction models for BCRP substrates.In this study, we developed support vector machine (SVM) models to predict wild-type BCRP substrates based on a total of 263 known BCRP substrates and non-substrates collected from literature. The final SVM model was integrated to a free web server.ResultsWe showed that the final SVM model had an overall prediction accuracy of ~73% for an independent external validation data set of 40 compounds. The prediction accuracy for wild-type BCRP substrates was ~76%, which is higher than that for non-substrates. The free web server (http://bcrp.althotas.com) allows the users to predict whether a query compound is a wild-type BCRP substrate and calculate its physicochemical properties such as molecular weight, logP value, and polarizability.ConclusionsWe have developed an SVM prediction model for wild-type BCRP substrates based on a relatively large number of known wild-type BCRP substrates and non-substrates. This model may prove valuable for screening substrates and non-substrates of BCRP, a clinically important ABC efflux drug transporter.

서포트벡터머신을 이용한 교육시설 초기 공사비 예측에 관한 연구

The accuracy of cost estimation at an early stage in school building project is one of the critical factors for successful completion. So various of techniques are developed to predict the construction cost accurately and expeditely. Among the techniques, Support Vector Machine(SVM) has an excellent ability for generalization performance. Therefore, the purpose of this study is to construct the prediction model for construction cost of educational building project using support vector machine technique. And to verify the accuracy of prediction model for construction cost. The performance data used in this study are 217 school building project cost which have been completed from 2004 to 2007 in Gyeonggi-Do, Korea. The result shows that average error rate was 7.48% for SVM prediction model. So using SVM model on predicting construction cost of educational building project will be a considerably effective way at the early project stage. 키워드 : 서포트벡터머신, 회귀분석, 공사비 예측, 교육시설

Application Study of Least Squares Support Vector Machines in Streamflow Forecast

In this article, the Least Square Support Vector Machine(LS-SVM) regression analysis and prediction methods were briefly introduced. Radial basis kernel function was chosen and a streamflow forecast model based on the toolbox of Matlab was constructed. Then the model was validated with a case study. After the model validation with a case study, it could be seen that the prediction model shows high accuracy and convergence speed. According to the analysis of the results, it is feasible for LS-SVM utilization in streamflow forecast.

Plasma Microrna Profile as a Predictor of Early Virological Response to Interferon Treatment in Chronic Hepatitis B Patients

Interferon (IFN) and pegylated interferon (PEG-IFN) treatment of chronic hepatitis B leads to a sustained virological response in a limited proportion of patients and has considerable side effects. To find novel markers associated with prognosis of IFN therapy, we investigated whether a pretreatment plasma microRNA profile could be used to predict early virological response to IFN. We performed microRNA microarray analysis of plasma samples from 94 patients with chronic hepatitis B who received IFN therapy. The microRNA profiles from 13 liver biopsy samples were also measured. The OneR feature ranking and incremental feature selection method were used to rank and optimize the number of features in the model. Support vector machine prediction engine and jack-knife cross-validation were used to generate and evaluate the prediction model. The optimized model consisting of 11 microRNAs yielded a 74.2% overall accuracy in the training group and was independently confirmed in the test group (71.4% accuracy). Univariate and multivariate logistic regression analyses confirmed its independent association with early virological response (OR=7.35; P=2.12×10(-5)). Combining the microRNA profile with the alanine aminotransferase level improved the overall accuracy from 73.4% to 77.3%. Co-transfection of an HBV replicative construct with microRNA mimics revealed that let-7f, miR-939 and miR-638 were functionally associated with the HBV life cycle. The 11 microRNA signatures in plasma, together with basic clinical variables, might provide an accurate method to assist in medication decisions and improve the overall sustained response to IFN treatment.

SVM Prediction of Ligand-binding Sites in Bacterial Lipoproteins Employing Shape and Physio-chemical Descriptors

Bacterial lipoproteins play critical roles in various physiological processes including the maintenance of pathogenicity and numbers of them are being considered as potential candidates for generating novel vaccines. In this work, we put forth an algorithm to identify and predict ligand-binding sites in bacterial lipoproteins. The method uses three types of pocket descriptors, namely fpocket descriptors, 3D Zernike descriptors and shell descriptors, and combines them with Support Vector Machine (SVM) method for the classification. The three types of descriptors represent shape-based properties of the pocket as well as its local physio-chemical features. All three types of descriptors, along with their hybrid combinations are evaluated with SVM and to improve classification performance, WEKA-InfoGain feature selection is applied. Results obtained in the study show that the classifier successfully differentiates between ligand-binding and non-binding pockets. For the combination of three types of descriptors, 10 fold cross-validation accuracy of 86.83% is obtained for training while the selected model achieved test Matthews Correlation Coefficient (MCC) of 0.534. Individually or in combination with new and existing methods, our model can be a very useful tool for the prediction of potential ligand-binding sites in bacterial lipoproteins.

BEST: Improved Prediction of B-Cell Epitopes from Antigen Sequences

Accurate identification of immunogenic regions in a given antigen chain is a difficult and actively pursued problem. Although accurate predictors for T-cell epitopes are already in place, the prediction of the B-cell epitopes requires further research. We overview the available approaches for the prediction of B-cell epitopes and propose a novel and accurate sequence-based solution. Our BEST (B-cell Epitope prediction using Support vector machine Tool) method predicts epitopes from antigen sequences, in contrast to some method that predict only from short sequence fragments, using a new architecture based on averaging selected scores generated from sliding 20-mers by a Support Vector Machine (SVM). The SVM predictor utilizes a comprehensive and custom designed set of inputs generated by combining information derived from the chain, sequence conservation, similarity to known (training) epitopes, and predicted secondary structure and relative solvent accessibility. Empirical evaluation on benchmark datasets demonstrates that BEST outperforms several modern sequence-based B-cell epitope predictors including ABCPred, method by Chen et al. (2007), BCPred, COBEpro, BayesB, and CBTOPE, when considering the predictions from antigen chains and from the chain fragments. Our method obtains a cross-validated area under the receiver operating characteristic curve (AUC) for the fragment-based prediction at 0.81 and 0.85, depending on the dataset. The AUCs of BEST on the benchmark sets of full antigen chains equal 0.57 and 0.6, which is significantly and slightly better than the next best method we tested. We also present case studies to contrast the propensity profiles generated by BEST and several other methods.

Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics

Excitation-emission matrices (EEM) and total synchronous fluorescence spectra (SFS) of normal and malignant breast tissue specimens are measured in UV-VIS spectral region to serve as data inputs in development of Support Vector Machine (SVM) based breast cancer diagnostics tool. Various input data combinations are tested for classification accuracy using SVM prediction against histopathology findings to discover the best combination regarding diagnostics sensitivity and specificity. It is shown that with EEM data SVM provided 67% sensitivity and 62% specificity diagnostics. With SFS data SVM provided 100% sensitivity and specificity for a several input data combinations. Among these combinations those that require minimal data inputs are identified.

Abstract 1722: Prognostic markers in early stage non small-cell lung cancer (NSCLC)

Abstract Conventional clinical and pathologic risk factors in stage I non small-cell lung cancer (NSCLC) provide limited prognostic information. Novel prognostic biomarkers are needed to identify patients with highest recurrence risk who will receive the greatest absolute risk reduction from adjuvant chemotherapy/radiation. Transcriptome, epigenome, and global genetic variation analyses of resected NSCLC frozen tumor samples (n=81; 29 stage 1A/52 stage 1B) were performed in order to develop an integrated prognostic signature. Only tumors that were resected within 3 months of the first biopsy-proven diagnosis were used to reduce lead time bias in timing of removal of the primary tumor. Here, the association of gene expression with disease free survival (DSF) is described. Whole genome expression data were quantile-normalized using the Illumina GenomeStudio 2011.1 software. Log-transformed data were then processed in Biometric Research Branch (BRB)-Array Tools. Prognostic models were developed with the top ranking genes using the supervised principal component survival algorithm (PCA). Risk group membership was then assigned based on this multivariate model using leave-one-out cross validation and class prediction modeling. Both models adjusted for age, gender and tumor histology. For stage 1A cases, a 113 gene signature selected by fitting a Cox proportional hazard model (Cox) was able to discriminate between poor and good DFS (α=0.001). Upon further analysis using a class prediction algorithm support vector machine (SVM), we identified an 18 gene classifier that predicted outcome SN=0.952, SP=0.571). A 10-fold cross validation ROC curve reported a value of 0.748. For stage 1B cases, the Cox model identified 141 genes (α=0.01), but the SVM prediction algorithm showed reduced SN/SP (0.75/0.25) for a 12 gene classifier, indicative of possible increased heterogeneity among the 1B cases. Nine genes overlapped between the PCA and class prediction model gene signatures. These nine are members of common lung tumorigenesis pathways, i.e. G-protein signaling regulation (KRAS/BRAF/RASSFI); tyrosine protein kinase signaling (EGFR/ERBB2); nucleotide excision repair (ERRC5 and ERCC6) and genes implicated in cell differentiation regulated by miR200B (HGD) or miR338 (TXNRD1), miRNAs independently found to be associated with prognosis in our cohort. These initial results demonstrate that a gene expression profile can distinguish stage I NCSLC tumors and predict prognoses, but require further validation. Supported in part by NIH 5P50 CA090440, P30CA047904, UPMC Institutional Funds. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1722. doi:1538-7445.AM2012-1722

Prediction of Safe Charge to Protect Surrounding Structures Using Support Vector Machine

The present paper mainly with deals the prediction of safe explosive charge used per delay (QMAX) using support vector machine (SVM) incorporating peak particle velocity (PPV) and distance between blast face to monitoring point (D). 150 blast vibration data sets were monitored at different vulnerable and strategic locations in and around a major coal producing opencast coal mines in India. 120 blast vibrations records were used for the training of the SVM model vis-a-vis to determine site constants of various conventional vibration predictors. Rest 30 new randomly selected data sets were used to compare the SVM prediction results with widely used conventional predictors. Results were compared based on coefficient of correlation (R) between measured and predicted values of safe charge of explosive used per delay (QMAX). It was found that coefficient of correlation between measured and predicted QMAX by SVM was 0.997, whereas it was ranging from 0.063 to 0.872 by different conventional predictor equations.

A Support Vector Machine Approach for the Prediction of Drilling Fluid Density at High Temperature and High Pressure

A support vector machine (SVM) approach was presented for predicting the drilling fluid density at high temperature and high pressure (HTHP). It is a universal model for water-based, oil-based, and synthetic drilling fluids. Available experimental data in the literature were used to develop and test this SVM model. Good agreement between SVM predictions and measured drilling fluid density values confirmed that the developed SVM model had good predictive precision and extrapolative features. The SVM model was also compared with the most popular models such as the artificial neural network (ANN) model, empirical correlations, and analytical models. Results showed that the SVM approach outperformed the competing methods for the prediction of drilling fluid density at HTHP.

Preprocessing unbalanced data using support vector machine

This paper deals with the application of support vector machine (SVM) to deal with the class imbalance problem. The objective of this paper is to examine the feasibility and efficiency of SVM as a preprocessor. Our study analyzes different classification algorithms that are employed to predict the customers with caravan car policy based on his/her sociodemographic data and history of product ownership. A series of experiments was conducted to test various computational intelligence techniques viz., Multilayer Perceptron (MLP), Logistic Regression (LR), and Random Forest (RF). Various standard balancing techniques such as under-sampling, over-sampling and Synthetic Minority Over-sampling TEchnique (SMOTE) are also employed. Subsequently, a strategy of data balancing for handling imbalanced distribution in data is proposed. The proposed approach first employs SVM as a preprocessor and the actual target values of training data are then replaced by the predictions of trained SVM. Later, this modified training data is used to train techniques such as MLP, LR, and RF. Based on the measure of sensitivity, it is observed that the proposed approach not only balances the data effectively but also provides more number of instances for minority class, which in turn enhances the performance of the intelligence techniques.