Model
A Model is created by applying an Algorithm with specific parameters on a training Dataset.
- Model URI is defined by dc:identifier
- Model Name is defined by dc:title
- Model creator might be defined by dc:creator
- The date of Model creation is defined by dc:date
- The Algorithm is defined by ot:algorithm
- The independent variables are Features and are defined by ot:independentVariables (can be multiple)
- The dependent variables are Features and are defined by ot:dependentVariables (can be multiple)
- The variables, where prediction results will be stored, are Features and are defined by ot:predictedVariables (can be multiple)
- Parameters are defined by ot:parameters
- The training Dataset is defined by ot:trainingDataset
- Example Model
<ot:Model rdf:ID="DemoModel"> <dc:creator rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >The model creator (perhaps a link to User resource)</dc:creator> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >The date of model creation</dc:date> <dc:format rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >The native format of the model content (e.g. PMML, Weka model, etc.)</dc:format> <dc:identifier rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >http://myservice/model/modelid</dc:identifier> <dc:title rdf:datatype="http://www.w3.org/2001/XMLSchema#string">Model name</dc:title> <ot:algorithm rdf:resource="#WekaJ48"/> <ot:dependentVariables rdf:resource="#MultiCellCall"/> <ot:independentVariables rdf:resource="#XLogPDescriptor"/> <ot:parameters> <rdf:Description> <rdf:type rdf:resource="http://www.opentox.org/api/1.1#Parameter"/> <ot:paramValue rdf:datatype="http://www.w3.org/2001/XMLSchema#string">-M 10</ot:paramValue> <ot:paramScope rdf:datatype="http://www.w3.org/2001/XMLSchema#string">optional</ot:paramScope> <dc:title rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >Max number of instances in a node</dc:title> </rdf:Description> </ot:parameters> <ot:parameters> <rdf:Description> <rdf:type rdf:resource="http://www.opentox.org/api/1.1#Parameter"/> <ot:paramValue rdf:datatype="http://www.w3.org/2001/XMLSchema#string">-P</ot:paramValue> <ot:paramScope rdf:datatype="http://www.w3.org/2001/XMLSchema#string">mandatory</ot:paramScope> <dc:title rdf:datatype="http://www.w3.org/2001/XMLSchema#string">Pruning</dc:title> </rdf:Description> </ot:parameters> <ot:predictedVariables rdf:resource="#MultiCellCallPredicted"/> <ot:trainingDataset rdf:resource="#DSSToxCPDBAS"/> </ot:Model>
Correction in the top bullet points
should be
The dependent variables are Features and are defined by ot:dependentVariables (can be multiple)