H-Tree

This is an implementation of the H-Tree Algorithm described in Chapter 21 'Classifying Streaming Data'
of Principles of Data Mining.

For the experiment with dataset lens24 described in Section 21.8.1 of Chapter 21 set
Number of Repetitions: 400 (use default values for other settings).

Dataset

Grace Period (records)

Probability for Use in Calculating Hoeffding Bound 0.9 0.95 0.99 0.999

Order of Inputting Records standard reverse random

Number of Repetitions of Training Data (minimum 100)

Use Hoeffding Bound yes no

acceptequals yes no

trace1   yes no    Main Trace Facility for Considering Splitting at a Node When Grace Count Achieved
trace2   yes no    Displays class/attval/total Values at All Expandable Leaf Nodes, When Grace Count is Reached at a Node
trace3   yes no    Display Class Totals at All Nodes, When Grace Count is Reached at a Node
trace4   yes no    Displays the Value of Information Gain For Each Available Attribute When Considering Expanding a Leaf Node
trace5   yes no    Displays Information about Branches and Nodes in the Classification Tree
trace6   yes no    When Grace Count is Reached at a Node, Displays Confusion Matrix of Predictions for All Records Since Previous Occurrence