/// <summary>
/// Method resposible for build up a tree
/// </summary>
/// <param name="attributeList">Array of attributes</param>
/// <param name="currentAttribute">Categorized Attribute</param>
/// <param name="data">Array of data</param>
/// <returns>A single node</returns>
private static TreeNode MountTree(shared.Attribute[] attributeList,
shared.Attribute categoricalAttribute,
string[][] data)
{
// if data is empty, return a single node with value failure
if (data.Length == 0)
return new TreeNode(new shared.Attribute(typeof(string), "Failure", -1, new string[] { "Failure" }));
var groupingData = data.GroupBy(m => m[categoricalAttribute.Index]);
/*
* if data consists of records all with the same value for
* the categorical attribute,
* return a single node with that value
*/
if (groupingData.Count() == 1)
return new TreeNode(new
shared.Attribute(
categoricalAttribute.Type,
categoricalAttribute.Name,
categoricalAttribute.Index,
new string[] { groupingData.First().Key }));
/*
* if attributeList is empty, the return a single node with as value
* the most frequent of the values of the categorical attribute
* that are found in the records of data
*/
if (attributeList.Length == 0)
return new TreeNode(GetMostFrequentValue(categoricalAttribute, data));
/*
* Let bestAttribute be the attribute with the largest gain
* among attributes in listAttributes
*/
shared.Attribute bestAttribute
= GetAttributeLargestGain(attributeList, categoricalAttribute, data);
// creates a root
TreeNode root = new TreeNode(bestAttribute);
// creates a new attribute list
shared.Attribute[] newAttributeList = attributeList
.Where(m => m.Name.Trim() != bestAttribute.Name.Trim()).ToArray();
// for each value in attributes
foreach (string value in root.Attribute.Values)
{
// if node type is continous range
if (root.Attribute.Type == typeof(double))
{
string[][] newData = null;
// gets the new data
if (value.IndexOf('<') > -1)
newData = data
.Where(m => double.Parse(m[bestAttribute.Index].Trim()) <
double.Parse(value.Trim().Replace("<", ""))).ToArray();
else
newData = data
.Where(m => double.Parse(m[bestAttribute.Index].Trim()) >=
double.Parse(value.Trim().Replace("=", "").Replace(">", ""))).ToArray();
// calls again
TreeNode childContinous = MountTree(newAttributeList, categoricalAttribute, newData);
root.Childs.Add(childContinous);
}
else // or node has a domain
{
// gets the new data
string[][] newData = data.Where(m => m[bestAttribute.Index].Trim() == value.Trim()).ToArray();
// calls again
TreeNode child = MountTree(newAttributeList, categoricalAttribute, newData);
root.Childs.Add(child);
}
}
return root;
}