The Programming Aided Instruction System Based On MOOC

With the promotion of deep integration of network technology and traditional education, Computer-Aided Instruction (CAI) has rapidly developed along with the arising of MOOC (Massive Open Online Courses) platform. The existing MOOC mainly provides students with teaching videos and exercises, but concerns few about the practice characteristics in programming training and personlized learning.In this thesis, we propose an integrated programming aided instruction modules for MOOC. The main research contents are listed as follows:(1) We propose the structuralization of teaching resources for MOOC. Compared with the lack of effective organization of teaching contents in traditional CAI, we propose the hierarchical curriculum contents of programming by means of knowledge-points tag-trees.(2) We propose the clustering solution for persionlized training for MOOC. To achieve personalized recommendation for different groups, The students are partitioned into different groups based on their basic information, learning degree and error knowledgep-points. Specically, the grouping procedure is automatically fulfilled by clusering algorithms.(3)We propose personalized teaching content recommendation algorithms for MOOC. Machine learning alorithms such as cooperration-filtering are employed for teaching resources distribution.(4) We put forward the whole teaching solution for MOOC integrated with face-to-face instrcution. Specfically, online evaluating, flow diagram aided instruction and error database etc modules are designed for computer aided instruction.The system can recommend different teaching resources to different students groups based on their characteristics, which can meet the study demand of different levels of students. Moreover, the hidden rules in the instructin is discovered which facilites the communication between teachers and students, both in the process of network and off-line teaching.