在非洲大陆，学生面临各种教育挑战，包括缺乏计算机、互联网连接、可靠电力等基本资源，以及合格教师短缺。尽管存在这些挑战，近期AI的进展（如BERT和GPT-4）已展现出推进教育的潜力。然而，这些AI工具往往主要在西方教育环境的背景下进行部署和评估，很少关注非洲学生面临的独特需求和挑战。在本章中，我们讨论了在非洲各地使用AI推进教育所面临的挑战。然后，我们描述了我们在非洲开发和部署AI教育工具以用于科学和计算教育的工作：（1）SuaCode，一款让非洲人能够使用智能手机学习编程的AI驱动应用；（2）AutoGrad，一款用于图形化和交互式编程作业的自动评分与反馈工具；（3）一款代码抄袭检测工具，可显示抄袭的视觉证据；（4）Kwame，一款面向编程课程的双语AI助教；（5）Kwame for Science，一款基于Web的AI助教，可即时回答学生的科学问题；（6）Brilla AI，一款参加全国科学与数学竞赛的AI选手。最后，我们讨论了利用AI推进非洲各地教育的潜在机遇。
英文摘要：
Across the African continent, students grapple with various educational challenges, including limited access to essential resources such as computers, internet connectivity, reliable electricity, and a shortage of qualified teachers. Despite these challenges, recent advances in AI such as BERT, and GPT-4 have demonstrated their potential for advancing education. Yet, these AI tools tend to be deployed and evaluated predominantly within the context of Western educational settings, with limited attention directed towards the unique needs and challenges faced by students in Africa. In this chapter, we discuss challenges with using AI to advance education across Africa. Then, we describe our work developing and deploying AI in Education tools in Africa for science and computing education: (1) SuaCode, an AI-powered app that enables Africans to learn to code using their smartphones, (2) AutoGrad, an automated grading, and feedback tool for graphical and interactive coding assignments, (3) a tool for code plagiarism detection that shows visual evidence of plagiarism, (4) Kwame, a bilingual AI teaching assistant for coding courses, (5) Kwame for Science, a web-based AI teaching assistant that provides instant answers to students' science questions and (6) Brilla AI, an AI contestant for the National Science and Maths Quiz competition. Finally, we discuss potential opportunities to leverage AI to advance education across Africa.
非洲的学生与教师的比例很高，限制了学生获得教师的机会。因此，学生们很难得到他们问题的答案。在这项工作中，我们扩展了我们以前的人工智能教学助手Kwame，将其调整为科学教育，并将其部署为Web应用程序。Kwame for Science基于西非高中证书考试（WASSCE）的综合科学科目回答学生的问题。Kwame for Science是一个基于句子BERT的问答网络应用程序，它显示3个段落作为答案，并在回答科学问题时显示信心分数。此外，它显示了前5个相关的过去考试的问题和他们的答案，除了3段。我们对Kwame for Science进行了为期2.5周的实际部署的初步评估，结果显示前3名的准确率为87.5%（n=56），共有11个国家的190名用户。Kwame for Science将为非洲数百万人提供可扩展的、具有成本效益的高质量远程教育。
Leveraging AI for Climate Resilience in Africa: Challenges, Opportunities, and the Need for Collaboration
Queen Mary University of London（伦敦玛丽女王大学）；ZYTLYN AfriClimate AI（ZYTLYN非洲气候人工智能）；McGill University（麦吉尔大学）；University of Rwanda（卢旺达大学）；African Institute for Mathematical Sciences（非洲数学科学研究所）；Google Research（谷歌研究院）；Karlstad University（卡尔斯塔德大学）
Generative AI for Education (GAIED): Advances, Opportunities, and Challenges
University of Auckland（奥克兰大学）；Microsoft（微软公司）；Worcester Polytechnic Institute（伍斯特理工学院）；EPFL（洛桑联邦理工学院）；Carnegie Mellon University（卡内基梅隆大学）；Carleton College（卡尔顿学院）；Max Planck Institute for Software Systems（马克斯·普朗克软件系统研究所）
Exploring Teachers' Perception of Artificial Intelligence: The Socio-emotional Deficiency as Opportunities and Challenges in Human-AI Complementarity in K-12 Education
Michigan State University（密歇根州立大学）；University of Pittsburgh（匹兹堡大学）
A Review of Generative AI in Computer Science Education: Challenges and Opportunities in Accuracy, Authenticity, and Assessment
School of Computer Science and Engineering, California State University, San Bernardino, USA（加州州立大学圣伯纳迪诺分校计算机科学与工程学院）；School of Information Systems and Technology, San Jose State University, USA（旧金山州立大学信息系统与技术学院）；University of California, San Diego, USA（加州大学圣地亚哥分校）
The Rise of Artificial Intelligence in Educational Measurement: Opportunities and Ethical Challenges
University of Alberta（阿尔伯塔大学）；Federation of American Scientists（美国科学家联合会）；Educational Testing Service（教育考试服务中心）；University of Maryland（马里兰大学）；University of Iowa（爱荷华大学）；Cambium Assessment（坎比姆评估公司）；Washington State University（华盛顿州立大学）；Acuity Insights（敏锐洞察公司）；PSI Services（PSI服务公司）；University of Delaware（特拉华大学）；Concordia University of Edmonton（埃德蒙顿康考迪亚大学）
