人工智能视域下机器学习的教育应用与创新探索

发布时间：2018-05-20 13:50

本文选题：机器学习 + 智慧教育　；参考：《大数据时代》2018年01期

【摘要】：新技术带来的教育变革方兴未艾,人工智能与智慧教育引领教育教学的创新,已经成为教育信息化发展的必然趋势。随着教育大数据的崛起,如何对大量数据进行分析以支持精准预测,是人工智能时代面对的一个新课题。机器学习作为人工智能的一个重要分支,能够满足教育大数据分析预测的需求。为此,基于"为何分析、分析什么、以何分析、何以应用"一系列问题,通过对机器学习的作用对象、作用过程、具体方法和利益相关者等方面的分析,探讨了机器学习和智慧教育的适切性。结合对近年来国外基于真实数据的机器学习教育应用案例研究成果的梳理和归纳,发现目前机器学习教育应用主要集中在学生建模、学生行为建模、预测学习行为、预警失学风险、学习支持和评测和资源推荐等六大方面。从跨界、技术和教学三个层面出发,基于智慧教育的框架对机器学习的教育应用与创新提出了相关建议。
[Abstract]:The educational reform brought by the new technology is in the ascendant. Artificial intelligence and wisdom education lead to the innovation of education and teaching, which has become the inevitable trend of the development of education information. With the rise of educational big data, how to analyze a large amount of data to support accurate prediction is a new subject in the era of artificial intelligence. As an important branch of artificial intelligence, machine learning can meet the demand of educational big data analysis and prediction. So, based on the "why analysis, what analysis, how to apply", through the analysis of the role of machine learning object, the process of action, specific methods and stakeholders, etc. The relevance of machine learning and intelligence education is discussed. By combing and summing up the case studies of machine learning education based on real data in recent years, it is found that the current application of machine learning education mainly focuses on student modeling, student behavior modeling, and prediction of learning behavior. Warning of school loss risk, learning support and evaluation and resource recommendation. Based on the framework of intelligent education, this paper puts forward some suggestions on the application and innovation of machine learning education from the three aspects of cross boundary, technology and teaching.
【作者单位】：华东师范大学教育信息技术学系;华东师范大学上海数字化教育装备工程技术研究中心;华东师范大学开放教育学院;
【基金】：全国教育科学“十二五”规划2014年度国家一般课题“智慧教育环境的构建与应用研究”(项目编号:BCA140051) 2017年度教育部在线教育研究基金(全通教育)课题“在线教育系统中学生反馈文本的情感分析技术与应用研究”(项目编号:2017YB126)的研究成果
【分类号】：G434

【相似文献】