Research And Implementation Of Data Analysis Technology In Scratch Programming Process

Scratch is a programming tool for children designed and developed by the Massachusetts Institute of Technology.The developers represented the programming instructions into blocks and realized visual programming.In China,Scratch is becoming the main tool for teaching teenagers the information technology.At present,many foreign scholars have noticed the importance of the Scratch programming process analysis and achieved some results.However,in China,Sctarch is only used for project creation,and the research on the process of creation is almost blank.In this thesis,we investigate the research methods and technologies of Scratch programming process at home and abroad,and propose a data extraction scheme based on Scratch programming action.On the basis of this scheme,we process a series of data analysis techniques based on time series,including time series-based operation mode clustering algorithm,time series-based coding habit extraction algorithm,and the correlation analysis of programming features with students and works.At the same time,we combined process data to improve the existing Scratch work scoring standards.On the basis of the existing Scratch programming platform,this paper reforms Scratch online editor and realizes the automatic extraction of Scratch process data.We also complete the work analysis module,not only realize the above data analysis algorithm,but also get the relationship between various types of programming features and students,as well as the relationship between Scratch works.Last but not least,we improve Scratch work scoring module to get more perfect scoring results.Through functional testing and system performance analysis,the test results are in line with expectations,which can and verify the scientificity and effectiveness of the algorithms proposed in this paper.This paper also verifies the accuracy of the new Scratch scoring criteria.