Data Mining Applied for Community Satisfaction Prediction of Rehabilitation and Reconstruction Project (Learn from Palu Disasters)

Abstract

Natural disasters can occur anytime and anywhere, especially in areas with high disaster risk. The earthquake that followed the tsunami and liquefaction in Palu, Indonesia, at the end of 2018 had caused tremendous damage. In recent years, rehabilitation and reconstruction projects have been implemented to restore the situation and accelerate economic growth. A study is needed to determine whether the rehabilitation and reconstruction that has been carried out for three years have met community satisfaction. The results of further analysis are expected to predict the level of community satisfaction for the implementation of rehabilitation and other reconstruction. The method used in this paper is predictive modeling using a data mining (DM) approach. Data were collected from all rehabilitation and reconstruction activities in Palu, Sigi, and Donggala with the scope of the earthquake, tsunami, and liquefaction disasters. The analysis results show that the Artificial Neural Network (ANN) and the support vector machine (SVM) with a DM approach can develop a community satisfaction prediction model to implement rehabilitation and reconstruction after the earthquake-tsunami and liquefaction disasters.

Implementation of Building Information Modelling for Road Rehabilitation and Reconstruction Project: Liquefaction Disaster of Palu Area

“…Nearly 30 minutes after the earthquake, a tsunami hit the coast of Palu City, Donggala Regency, and surrounding areas along the coast, followed by landslides and Liquefaction in mountainous and urban areas. These events caused significant damage to infrastructures such as roads, bridges, airport buildings, water supplies, and other public facilities, including hospitals and schools (Irfan Rifai, 2022). The catastrophic events that co-occurred became a natural Page │782 laboratory on a full scale.…”