SUMMARY

• Produced comparable results as the Bayesian approach on the scenic beauty estimation and surname pronunciation generation problems

  
  

• Presented the mathematical framework used by the C4.5 decision tree algorithm

  
  

• Implemented C4.5 decision tree:

  
  

  • Splitting rules include information gain, gain ratio, and a combination of both

    
    

  • Pessimistic pruning method

    
    

• Developed a decision tree software package which can handle large data sets with large number of classes and attributes, as well as special characters for class and attribute values

  
  

• Software is available from the ISIP web site at: /hse/ies/resources/technology/software/1998/decision_tree