private DecisionTree decisionTreeLearning(DataSet ds, ICollection <string> attributeNames, ConstantDecisonTree defaultTree)
{
if (ds.size() == 0)
{
return(defaultTree);
}
if (allExamplesHaveSameClassification(ds))
{
return(new ConstantDecisonTree(ds.getExample(0).targetValue()));
}
if (attributeNames.Size() == 0)
{
return(majorityValue(ds));
}
string chosenAttribute = chooseAttribute(ds, attributeNames);
DecisionTree tree = new DecisionTree(chosenAttribute);
ConstantDecisonTree m = majorityValue(ds);
ICollection <string> values = ds.getPossibleAttributeValues(chosenAttribute);
foreach (string v in values)
{
DataSet filtered = ds.matchingDataSet(chosenAttribute, v);
ICollection <string> newAttribs = Util.removeFrom(attributeNames, chosenAttribute);
DecisionTree subTree = decisionTreeLearning(filtered, newAttribs, m);
tree.addNode(v, subTree);
}
return(tree);
}
//
// PRIVATE METHODS
//
private DecisionTree DecisionTreeLearning(DataSet ds,
IList <string> attributeNames, ConstantDecisonTree defaultTree)
{
if (ds.Count == 0)
{
return(defaultTree);
}
if (this.AllExamplesHaveSameClassification(ds))
{
return(new ConstantDecisonTree(ds.GetExample(0).TargetValue()));
}
if (attributeNames.Count == 0)
{
return(this.MajorityValue(ds));
}
string chosenAttribute = this.ChooseAttribute(ds, attributeNames);
var tree = new DecisionTree(chosenAttribute);
ConstantDecisonTree m = this.MajorityValue(ds);
IList <string> values = ds.GetPossibleAttributeValues(chosenAttribute);
foreach (string v in values)
{
DataSet filtered = ds.MatchingDataSet(chosenAttribute, v);
IList <string> newAttribs = Util.Util.RemoveFrom(attributeNames,
chosenAttribute);
DecisionTree subTree = this.DecisionTreeLearning(filtered, newAttribs, m);
tree.AddNode(v, subTree);
}
return(tree);
}