[4G1-OS-4a] 人工知能におけるプライバシー，公平性，説明責任，透明性への学際的アプローチ(1/2)

[4G1-OS-4a-03] Discussion on a definition of diversity with similarity and networks

〇Keita Kinjo¹ (1. Kyoritsu Women's University, Faculty of Business)

Keywords:diversity, inclusion, network, fairness, AI ethics

Diversity and inclusion are attracting attention in society. These are also attracting attention in the fundamentals and applications of artificial intelligence, including machine learning. In this paper, we propose a new diversity index called diversity index with similarity and network. The proposed index will enable us to analyze the detailed structure of diversity and inclusiveness, which were not focused. To validate this index, we surveyed people’s preferences for the rate of people and network structure in an organization and examined the correlation between the index and their preferences. Consequently, a high correlation between them was confirmed. The proposed index will be useful for designing artificial intelligence with diversity and inclusion.