Dini Destiani Siti Fatimah
Institut Teknologi Garut

Published : 147 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Jurnal Teknik Informatika (JUTIF)

C4.5 ALGORITHM FOR DISASTER IDENTIFIER SYSTEM Ade Sutedi; Hilmi Aulawi; Eko Walujodjati; Dini Destiani Siti Fatimah
Jurnal Teknik Informatika (Jutif) Vol. 3 No. 3 (2022): JUTIF Volume 3, Number 3, June 2022
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20884/1.jutif.2022.3.3.160

Abstract

Disaster management is a strategic issue that has been widely studied as a form of mutual responsibility in reducing victims and losses due to disasters. Today, one of the sources of disaster information is spread on Twitter social media. This research explains the implementation of the C4.5 algorithm to classify and map the disaster information that delivers using social media Twitter from the official BNPB_Indonesia account. The disaster data was retrieved and then processed to display in the geographic information system form. The words combination of disaster events, victims, and disaster locations process is carried out using n-gram by divided into unigram, bigram, and trigram to obtain the vocabulary accordance with the database. The C4.5 algorithm in this research was used to classify the disaster information with several categories. The results shows that the C4.5 algorithm can be used to classified the category of disaster and could identified the disaster informasion such as type of disaster, victims, and locations. The result can provide real time information on the distribution of disaster events and their locations using geographical information system. However, for location such as name of provinces only which has many geo-position possibilities (district or sub-district). The determination of the disaster location could be difficult. In addition, to determine the information obtained from post-disaster conditions such as the number of victims, damage, and losses. The comparison of n-gram with predetermined keywords is still constrained by noise of data.