Abstract—Efficient data manipulation and retrieval is a
fundamental part of many business processes in the majority of
todays’ companies. SQL, as a standard, is widely adopted and
well accepted in this area. Students who set out to learn SQL
frequently face difficulties. The learning process is to some
extent inefficient, as the student’s knowledge is afterwards often
inadequate. Several computer-aided systems have been
developed to alleviate the problem. However, most of them are
static and rigid, because the system’s knowledge is encoded
manually. We propose a new system based on past attempts and
solutions to SQL exercises. The proposed system is flexible and
dynamic, as it adapts to the individual student and requires
minimal intervention from domain experts. We show that the
system is beneficial, in particular to students with low prior
knowledge.
Index Terms—Intelligent tutoring systems, SQL learning,
Markov Decision Processes, adaptive hint generation.
Tadej Matek is with the Faculty of Computer and Information Science,
University of Ljubljana, Slovenia.
Aljaž Zrnec and Dejan Lavbič are with the Laboratory for Data
Technologies, Slovenia (e-mail: Dejan.Lavbic@fri.uni-lj.si).
Cite: Tadej Matek, Aljaž Zrnec, and Dejan Lavbič, "Learning SQL with Artificial Intelligent Aided Approach," International Journal of Information and Education Technology vol. 7, no. 11, pp. 803-808, 2017.