Monthly Groundwater Level Modeling Using Data Mining Approaches.

Abstract

Determination of the fluctuations in groundwater level (GWL) in terms of planning and operating their resources is important. In Turkey, many basins are experiencing problems in terms of the potential of groundwater. Increasing water demand, adverse conditions created by climate change and lack of planning related to underground water management in the basin have increased these problems. As a field of application, it was applied for General Directorate of State Hydraulic Works (DSI) well of Hatay province in Turkey. In the study, GWL predictions were evaluated using data mining approaches such as Radial Basis Neural Network (RBNN) and Support Vector Machines (SVM) methods. Monthly data sets between 2002 and 2015, including hydrological parameters predict the GWL used. According to comparison results, it was observed that the data mining models gave good results for observation in test phase.

Keywords

• ground water level
• prediction
• neural network
• support vector machines
• data mining.