GIS-Based Analysis of Changing Surface Water in Rajshahi City Corporation Area Using Support Vector Machine (SVM), Decision Tree & Random Forest Technique

Abstract

Water is one of the essential natural resources of nature. All living creature depends on water. Living creatures are using water for their different purposes. Earth’s large portion is covered by salt water but very less has fresh water. Freshwater can be found as groundwater and surface water. Surface water is stored as waterbodies on the surface of this world. Ponds, canals, rivers, and lakes are some of the waterbodies that provide fresh water to us. These waterbodies are fulfilling our need for fresh water. Most of the waterbodies are drying up for natural disasters or they are continuously filling by humans. These resources need some of our attention to preserve them. Rajshahi Development Authority (RDA) and United States Geological Survey (USGS) provide important data for this research. Waterbodies are detected by using Geographic Information System (GIS), GIS gives us the power of mapping and store, detect, and manipulate spatial or geographic data. Images are collected from the Landsat 4-5 Thematic Mapper (TM) and Landsat 8 Operational Land Imager (OLI). They are classified by using ArcGIS. Images are classified in maximum likelihood classification by generating signature files to extract feature. Percentage of waterbodies in each year is calculated from the attribute table. A dataset is prepared from these features and tested on different classification techniques. Support Vector Machine (SVM), Decision Tree and Random Forest Technique are implemented on this dataset. Among them, Random Forest shows 92% accuracy, which is better from other techniques. These algorithms also measure the precision, recall, and f1 scores of the classifiers. The precision, recall, and f1-score of random forest technique show 0.943, 0.920, 0.922, which indicate better accuracy than other techniques. Keywords: