カンファレンス (国内) 災害時の時々刻々と変わる情報ニーズの把握
坪内 孝太, 日暮 立, 藤島新也 (日本放送協会), 麻生重樹 (日本放送協会)
2023年度 人工知能学会全国大会（第37回） (JSAI 2023)
This study, utilizing past actual disasters as case studies, analyzed the location and search history information of users in affected areas to determine the existence of specific evacuation signals in each area. The data from the 2022 Typhoon No. 14 disaster was evaluated, and information regarding the operation of major transportation systems and the closure of local grocery stores were identified as evacuation signals. Regional variations in evacuation signals were also identified. Providing information that is pertinent to the user’s context can elicit appropriate evacuation behavior in the event of a disaster.