{"title":"SVM Learner","nodeType":"Learner","id":"*eTTAGRT-ku7Z8LJb","factoryName":"org.knime.base.node.mine.svm.learner.SVMLearnerNodeFactory2","factoryId":"org.knime.base.node.mine.svm.learner.SVMLearnerNodeFactory2","path":["Analytics","Mining","SVM"],"owner":"knime","ownerAccountId":"account:team:b5f0848f-496d-4e6a-b869-fa3532248db7","description":"<p>This node trains a support vector machine on the input data. It supports a number of different kernels (HyperTangent, Polynomial and RBF). The SVM learner supports multiple class problems as well (by computing the hyperplane between each class and the rest), but note that this will increase the runtime.</p>\n            \n            <p>The SVM learning algorithm used is described in the following papers: <a href=\"https://www.microsoft.com/en-us/research/publication/%20fast-training-of-support-vector-machines-using-sequential-minimal-optimization/\" rel=\"ugc nofollow noopener noreferrer\">Fast Training of Support Vector Machines using Sequential Minimal Optimization</a> , by John C. Platt and <a href=\"https://digilander.libero.it/sedekfx/papers_/smo_mod.pdf\" rel=\"ugc nofollow noopener noreferrer\"> Improvements to Platt&#39;s SMO Algorithm for SVM Classifier Design</a> , by S. S. Keerthi et. al.</p>\n            \n        \n        ","shortDescription":"Trains a support vector machine.","keywords":["support vector machine"],"deprecated":false,"streamable":false,"tags":["Analytics","Mining","SVM"],"icon":"data:image/png;base64,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","views":[{"name":"SVM View","description":"Shows the trained Support Vector Machines for each class with their corresponding support vectors."}],"inPorts":[{"objectClass":"org.knime.core.node.BufferedDataTable","color":"000000","dataType":"Table","optional":false,"name":"Training Data","description":"Datatable with training data."}],"outPorts":[{"objectClass":"org.knime.core.node.port.pmml.PMMLPortObject","color":"1469af","dataType":"PMML","optional":false,"name":"Trained SVM","description":"Trained Support Vector Machine."}],"dynInPorts":[],"dynOutPorts":[],"additionalSiteInformation":{"url":"https://update.knime.com/analytics-platform/5.12","trusted":true,"enabledByDefault":true,"name":"KNIME Analytics Platform 5.12 Update Site","version":"5.12.0"},"dialog":[{"sectionName":"Options","fields":[{"name":"Class column","description":"Choose the column that contains the nominal target variable.","optional":false},{"name":"Overlapping penalty","description":"The overlapping penalty is useful in the case that the input data is not separable. It determines how much penalty is assigned to each point that is misclassified. A good value for it is 1.","optional":false},{"name":"Kernel type","description":"Choose the kernel function for the support vector machine. Each kernel has its own parameters which can be configured below.\n<ul><li><b>Polynomial</b> : Polynomial kernel of the form (gamma * x * y &#43; bias)^power.</li><li><b>HyperTangent</b> : Hyperbolic tangent kernel of the form tanh(kappa * x * y &#43; delta).</li><li><b>RBF</b> : Radial Basis Function kernel of the form exp(-||x - y||^2 / (2 * sigma^2)).</li></ul>","optional":false},{"name":"Bias","description":"The bias parameter for the polynomial kernel.","optional":false},{"name":"Power","description":"The power parameter for the polynomial kernel.","optional":false},{"name":"Gamma","description":"The gamma parameter for the polynomial kernel.","optional":false},{"name":"Kappa","description":"The kappa parameter for the hyperbolic tangent kernel.","optional":false},{"name":"Delta","description":"The delta parameter for the hyperbolic tangent kernel","optional":false},{"name":"Sigma","description":"The sigma parameter for the Radial Basis Function (RBF) kernel.","optional":false}]}],"links":[],"kudos":[],"kudosCount":0,"bundleInformation":{"bundleVersion":"5.12.0.v202606261225","bundleName":"KNIME Base Nodes","bundleSymbolicName":"org.knime.base","bundleVendor":"KNIME AG, Zurich, Switzerland","featureSymbolicName":"org.knime.features.base","featureName":"KNIME Base nodes","featureVersion":"5.12.0.v202606261225","featureVendor":"KNIME AG, Zurich, Switzerland","extensionId":"*bjR3r1yWOznPIEXS"},"frequentSuccessors":[{"id":"*z4QawMJlTQW3cArl","factoryName":"org.knime.base.node.mine.svm.predictor2.SVMPredictorNodeFactory","probability":0.9457286432160804},{"id":"*14A00VB0oh4by3NE","factoryName":"org.knime.base.node.io.pmml.write.PMMLWriterNodeFactory","probability":0.009045226130653266},{"id":"*UwpLkzO3YUH9ogjZ","factoryName":"org.knime.base.node.io.predictor.PredictorWriterNodeFactory","probability":0.006030150753768844},{"id":"*W_FO2BWH7C8nF7au","factoryName":"org.knime.ensembles.pmmlporttocell.PMMLPortToCellNodeFactory","probability":0.006030150753768844},{"id":"*mTvdLptC3aqDqGrI","factoryName":"org.knime.base.node.mine.decisiontree2.predictor2.DecTreePredictorNodeFactory","probability":0.004020100502512563},{"id":"*Z_yB8Lpw103-H_Dc","factoryName":"org.knime.base.pmml.transformation.PmmlPreprocessorNodeFactory","probability":0.003015075376884422},{"id":"*a36eXM94uKOrTIlo","factoryName":"org.knime.ensembles.boosting.BoostingLearnerLoopEndNodeFactory","probability":0.003015075376884422},{"id":"*GWhtZpnrDMNapedT","factoryName":"org.knime.ensembles.modeltotable.ModelToTableNodeFactory","probability":0.003015075376884422},{"id":"*lISJ_edcE_TG6MVp","factoryName":"org.knime.base.node.mine.bayes.naivebayes.predictor3.NaiveBayesPredictorNodeFactory2","probability":0.003015075376884422},{"id":"*NYPEQwQdVg09dFhS","factoryName":"org.knime.ensembles.pmmlpredict2.PMMLPredictor2NodeFactory","probability":0.003015075376884422}]}