A framework for efficient decision tree using priority chunk learning

In data mining, decision trees are one of the very efficient techniques for knowledge discovery. Decision trees are very popular due to easy intercept and understanding. One of the shortcomings for decision trees is the huge data available for analysis. The process of knowledge discovery can be break down into different phases for efficient knowledge discovery. In this paper, we propose a novel Priority Chunk Learning (PCL) technique for efficient knowledge discovery from the varied datasets. We extensively evaluated the performance of the PCL approach o the 14 UCI datasets. The experimental results show that the proposed PCL approach has performed better than the benchmark algorithms.