Data mining applied for liquefaction mapping and prediction learn from palu earthquakes

Abstract

The 2018 Palu-Sigi-Donggala earthquakes in Center Celebes have caused significant damage to many residential houses due to varying degrees of soil liquefaction over a vast extent of urban areas unseen in past destructive earthquakes. Soil liquefaction occurred in Palu and Sigi, thus providing researchers with a wide range of characterizing soil and site response to large-scale earthquake shaking. One of the essential learning issues is the prediction of liquefaction. Prediction of liquefaction is also a complex problem as it depends on many different physical factors, and the relations between these factors are highly non-linear and complex. Most of these approaches are based on classical statistical criteria and neural networks. In this paper, a new method which is based on classification data mining (DM) is proposed. The proposed approach is based on historical data from the field and sciences portal. The proposed algorithm is also compared with several other DM algorithms based on the miner. It is shown that the proposed algorithm is very useful and accurate in the prediction of liquefaction.