Biography: Shinichi Oeda is an Associate Professor of National Institute of Technology, Kisarazu College, Japan. He received a Ph.D. in Engineering from University of Tokyo Metropolitan Institute of Technology in 2004. He was a research associate, an assistant professor, a senior lecturer and an associate professor at National Institute of Technology, Kisarazu College from 2004 to 2017. His research interests include artificial intelligence, machine learning, educational data mining and healthcare support system. He is a member of IPSJ and IEICE.
Speech Title: Extracting Latent Skills from Examinations Results and Log-Data in Programming Classes
Abstract: Intelligent tutoring systems and learning management systems have been widely used in the fields of education, and have allowed us to collect log data from learners, such as students. Educational data mining (EDM) aims at discovering useful information from the massive amounts of electronic data collected by these educational systems. EDM involves the application of data mining, machine learning, and statistics to information generated from educational settings. Examinations are tools for measuring examinees&amp;#39; skills. A question item in an examination requires several skills to solve it. It is important to find which skills an item requires in order to comprehend latent skills. The relationship between items and skills can be represented by a Q-matrix. Recent studies have attempted to extract a Q-matrix by non-negative matrix factorization (NMF) from a set of examinees&amp;#39; test scores. However, conventional studies including us have not used log-data during learning. In this paper, we develop to extract the latent skills from examination results and log-data.