リ カンコウ   Guanhong LI   李　冠宏    所属   京都外国語大学  国際貢献学部 グローバル観光学科    職種   講師
発表年月日	2024/05/29
発表テーマ	An innovative approach to studying complex social problems by combining multi-agent simulation and model-checking
会議名	The 38th Annual Conference of the Japanese Society for Artificial Intelligence
学会区分	全国学会
発表形式	口頭（一般）
単独共同区分	単独
開催期間	2024/05/28～2024/05/31
概要	This study addresses the challenges of analyzing complex social problems in the context of multi-agent simulation. Computational social scientists often face difficulties in effectively measuring issues such as sustainability and creativity. These issues involve dynamic processes that are difficult to represent by a single state or indicator. To improve the analysis of these dynamic social problems, we propose the use of model checking, a formal verification technique commonly used in hardware and software system design. By representing social problems as transition systems, each state defined by multiple indicators, and formulating social properties in a formal language, model checking tools can verify their satisfiability under certain conditions. We illustrate the potential of this approach with a multi-agent model based on opinion dynamics, demonstrating its applicability to the study of complex social phenomena. This integration of model checking into socio-dynamic system modeling offers a innovative avenue for advancing computational social science research on diverse topics.