A Regression Tree Application on Soil and Road Dust Pollution

• Activity Number: 428
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Betül Kan*+ and Berna Yazici and Semra Malkoç and Metin Altan and Savas Koparal
• Companies: Anadolu University and Anadolu University and Anadolu University and Anadolu University and Anadolu University
• Keywords: Classification and Regression Analysis ; Regression Trees ; Splitting ; Heavy Metals ; Soil ; Road Dust

Abstract

In this study, an iterative algorithm of splitting data into partitions, regression tree, is applied to explore the soil pollution at different sample points. The algorithm tries breaking up both for the soil and road dust measurements, using every possible binary split on every sample point. The measurements of eight heavy metals of soil and road dust are taken into consideration as outputs at some traffic density points where both tramway and roadway are and only roadway or tramway is in use. The difference in soil and road dust pollution measurements between different sample points are discussed and analyzed which may provide some useful information for further environmental applications. Also the pollution maps regarding the results of regression trees are plotted as digital maps by using geographical information systems techniques for those points that are taken into account.