Extracting structure of latent skills from examination results in programming education

2015 Fiscal Year Final Research Report

• Project/Area Number: 25750095
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Educational technology
• Research Institution: Kisarazu National College of Technology
• Principal Investigator: Oeda Shinichi 木更津工業高等専門学校, その他部局等, 准教授 (80390417)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 非負値行列因子分解 / Educational Data Mining / プログラミング教育

Outline of Final Research Achievements

Recently, ITS(Intelligent Tutoring Systems) has been widely used, these systems has allowed the collection of huge amounts of educational data, such as examination results or studying behavior of students. In this study, we extract latent skills to master programming technique from the point of view of EDM (Educational Data Mining).
Our studies applied NMF(non-negative matrix factorization) method to decompose the results of an examination into a Q-matrix and another matrix. The Q-matrix is the relationship between
items and skills. Our proposed method could extract and visualize latent skills of students.

Free Research Field

知能システム，教育工学，機械学習，データマイニング