D-Loc Apps: A Location Detection Application Based on Social Media Platform in the Event of A Flood Disaster

Abstract

The purpose of this research is to develop a disaster location detector application, specifically about flood that happens in the area of Jakarta, Indonesia. According to social media data updated by the users of twitter, search inquiry is done by its users with the keyword 'seputaran banjir' (in Bahasa Indonesia) which is combined with the Location Based Service function of the determined word search. The method in creating this application is Extreme Programming that uses repetitive and incremental development process approach. By using twitter status data about Jakarta's area, this application has been successfully developed and the result can be shown in the form of twitter status updates from its users that contain words about flood. The search results of flood location validation are coordinates, which are done by checking the availability of the geospatial information from each update of the data. The average accuracy of the search result is 83.7% out of all the location search results.