Analysis of Sequential Order Incremental Methods in Predicting the Number of Victims Affected by Disasters. (August 2019)

• Record Type: Journal Article
• Authors: Parulian, Parulian; Tinambunan, Medi Hermanto; Ginting, Salomo; Khalil Gibran, M.; Wanto, Anjar; Muharram, La Ode; Nurmawati, N; Bhawika, Gita Widi
• Is Part Of: Journal of physics. Volume 1255(2019)
• Publication Date: 2019-08
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1255/1/012033
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Disaster is a series of events that threaten and disrupt human life caused by natural factors, non-natural factors and human factors themselves. Therefore, disasters cause casualties, environmental damage, property losses, and psychological impacts. In this study will be discussed about the prediction of the number of victims affected by the disaster, either died, lost, injured, suffered or displaced. Data sources were obtained by the National Disaster Management Agency and the Indonesian Central Statistics Agency. The method used to predict is the Incremental Sequential Order method. This method is one part of the Artificial Neural Network method. With this method, network architecture patterns will be established to predict the number of victims affected by the disaster for years to come. The network architecture models used are 4-5-1, 4-10-1, 4-5-10-1, 4-10-20-1 and 4-15-30-1. Of the five models, the best models will be obtained, namely 4-15-30-1 with an accuracy rate of 80%. With this architectural model, predictions will be made on the number of victims affected by the disaster for years to come.