Parallel Decision Tree with Application to Water Quality Data Analysis

Decision tree is a popular classification technique in many applications, such as retail target marketing, fraud detection and design of telecommunication service plans. With the information exploration, the existing classification algorithms are not good enough to tackle large data set. In order to deal with the problem, many researchers try to design efficient parallel classification algorithms. Based on the current and powerful parallel programming framework -- MapReduce, we propose a parallel ID3 classification algorithm(PID3 for short). We use water quality data monitoring the Changjiang River which contains 17 branches as experimental data. As the data are time series, we process the data to attribute data before using the decision tree. The experimental results demonstrate that the proposed algorithm can scale well and efficiently process large datasets on commodity hardware.