Host: The Japanese Society for Artificial Intelligence
Name : 34th Annual Conference, 2020
Number : 34
Location : Online
Date : June 09, 2020 - June 12, 2020
Recently it becomes more siginificant on evacuation simulation to build safety plans which minimizes disaster damages. Especially analysis on group behavior characteristics is important clue to have evacuees with unexpected behaviors or identify risks such as "falling one after another" in congession. Nevertheless research works have not been established concerning group behavior itself. This paper addresses analysis of group behavior characteristics in evacuation using "topic model" towards multi-agent evacuation simulation log, especially focusing on followee-follower action.