Data_Sheet_1_Qualitative and quantitative analyses of artificial intelligence ethics in education using VOSviewer and CitNetExplorer.ZIP

The new decade has been witnessing the wide acceptance of artificial intelligence (AI) in education, followed by serious concerns about its ethics. This study examined the essence and principles of AI ethics used in education, as well as the bibliometric analysis of AI ethics for educational purposes. The clustering techniques of VOSviewer (n = 880) led the author to reveal the top 10 authors, sources, organizations, and countries in the research of AI ethics in education. The analysis of clustering solution through CitNetExplorer (n = 841) concluded that the essence of AI ethics for educational purposes included deontology, utilitarianism, and virtue, while the principles of AI ethics in education included transparency, justice, fairness, equity, non-maleficence, responsibility, and privacy. Future research could consider the influence of AI interpretability on AI ethics in education because the ability to interpret the AI decisions could help judge whether the decision is consistent with ethical criteria.

近十年来，人工智能（Artificial Intelligence，AI）在教育领域获得了广泛应用，但与此同时，其伦理问题也引发了诸多严肃担忧。本研究针对教育场景下人工智能伦理的核心内涵与原则展开了系统探讨，同时对面向教育用途的人工智能伦理相关研究开展了文献计量分析。本研究借助VOSviewer的聚类分析技术（样本量n=880），梳理出了教育人工智能伦理研究领域排名前十的作者、文献来源、研究机构与国家。本研究通过CitNetExplorer对聚类结果进行分析（样本量n=841）后得出结论：面向教育场景的人工智能伦理核心内涵包含义务论、功利主义与德性论；而教育领域人工智能伦理的原则则包括透明度、正义、公平、平等、不伤害、责任与隐私。未来研究可围绕人工智能可解释性对教育领域人工智能伦理的影响展开探索，因为对人工智能决策进行解释的能力，有助于判断该决策是否符合伦理准则。