net.sf.jdmf.algorithms.classification
Class OneRuleAlgorithm

java.lang.Object
  net.sf.jdmf.algorithms.AbstractDataMiningAlgorithm
      net.sf.jdmf.algorithms.classification.OneRuleAlgorithm

All Implemented Interfaces:

DataMiningAlgorithm

public class OneRuleAlgorithm
extends AbstractDataMiningAlgorithm

Implements the 1R classification algorithm as described in Data Mining: Practical Machine Learning Tools and Techniques (Second Edition) by Ian H. Witten and Eibe Frank (Morgan Kaufmann, 2005).

 
 The idea is simple:
 1. Take an attribute.
 2. Select a value of this attribute.
 3. Select a value of the decision attribute.
 4. Calculate the coverage of an instance containing both attribute values.
 5. Repeat 3-4 for all decision values.
 6. Choose the instance with the highest coverage (randomly, if there are 
 more winners) and convert it to a rule.
 7. Repeat 2-6 for all values of this attribute. 
 8. Calculate the error sum for all rules based on this attribute. The error
 rate for each rule is defined as the number of times the attribute value 
 occurs minus the coverage of the rule.
 9. Repeat 1-8 for all attributes.
 10. Choose the attribute with the smallest error sum and output 
 classification rules based on this attribute.
 
 

If the attribute is numeric, the values are partitioned to achieve a reasonable number of rules (instead of calculating coverage, which would result in quite a few rules...).

Author:

quorthon

See Also:

AttributeValuePartitioner

Constructor Summary

OneRuleAlgorithm()

Method Summary

DataMiningModel

analyze(InputData inputData) Analyzes input data (attributes and decisions) and produces output data (rules, decision trees, clusters, ...).

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

OneRuleAlgorithm

public OneRuleAlgorithm()

Method Detail

analyze

public DataMiningModel analyze(InputData inputData)

Description copied from interface: DataMiningAlgorithm

Analyzes input data (attributes and decisions) and produces output data (rules, decision trees, clusters, ...).

See Also:

DataMiningAlgorithm.analyze(net.sf.jdmf.data.input.InputData)