Multiclass Multifeature Split Decision Tree Construction in a Distributed Environment
Jie Ouyang, Nilesh V. Patel, Ishwar K. Sethi · 2008
The decision tree-based classification is a popular approac h for pattern recognition and data mining. Most decision tree induction methods assume training data being present at one central location. Given the growth in distributed databases at geographically dispersed locations, the methods for decision tree inducti on in distributed settings are gaining importance. This paper descr ibes one such method that generates compact trees using multifeature splits in place of single feature split decision trees generated by most existing methods for distributed data. Our method is based on Fisher’s linear discriminant function, and is capable of dealing wit h multiple classes in the data. For homogeneously distributed data, the decision trees produced by our method are identical to decision trees generated using the Fisher’s linear discriminant fun ction with centrally stored data. For heterogeneously distributed da ta, a certain approximation is involved with a small change in performance with respect to the tree generated with centrally stored dat a. Experimental results for several well known data sets are presented and compared with decision trees generated using the Fisher’s l inear discriminant function with centrally stored data.