基于教育大数据的量化自我MOOC自适应学习系统研究

Research on Quantified Self MOOC Adaptive Learning System Based on Educational Big Data

教育大数据创新性地完善了当前学习系统的架构,实现了基于数据流的学习分析和挖掘机制,让以往难以实现的精准分析成为了可能,而量化自我算法将成为教育大数据分析和实现自适应学习的关键所在。本文首先分析了教育大数据背景下作为个人级数据应用的量化自我概念。然后,讨论如何通过全面地记录、跟踪和可视化学习者的学习行为,促使量化自我算法更容易、准确地获得学习者的经验,实现以学习者的认知需求为中心来优化学习者的学习过程。进而提出基于量化自我算法的MOOC自适应学习系统的模型,并且对该模型的结构进行了详细分析。最后,结合基于网络学习行为分析的智能反馈策略和认知思维层次的在线学习行为分类,构建了量化自我学习算法QSLA(Quantified Self Learning Algorithm)作为实现基于教育大数据的自适应学习的基础。

Educational big data have creatively improved and perfected the architecture of learning system and realized learning analytics and data mining based on data flow. Educational big data thus have also made it possible to achieve precise analysis of learning, which was difficult in the past. Quantified self algorithm will become the key to the analysis of educational big data and the application of adaptive learning system. This paper first analyzes the concept of quantified self the use of educational big data at individual learner level. Then, this paper discusses how to utilize comprehensive recording, tracking, and visualization of learners＂ behavior to assist quantitative self algorithm to acquire learners＂ experiences more easily and accurately, and thus to optimize learners＂ learning process centered on the their cognitive needs. The paper further proposes a model of MOOC adaptive learning system based on quantified self algorithm and explains the structure of this model in detail. Finally, based on the intelligent feedback strategy and the cognitive level of classification of online learning behavior, this paper advances the QSLA （Quantified Self Learning Algorithm） as the foundation to realize adaptive learning based on educational big data.