Mining MySQL error logs to map student learning

Abstract

Information management (IM) is an undergraduate program related to information technology education (ITE) that focuses on database design and management. However, reports reveal that there is a significant skills shortage in ITE graduates along database analysis and design, and academia is being unresponsive to solve the skill gap. This research intends to corroborate the students' manner of learning the SQL language with the knowledge units (KUs) that students find difficult in the IM course and recommend interventions to overcome the skill deficiency. The study used mixed methods where the collection of data was administered through an application program that collects MySQL logs to generate analysis using naïve Bayes and Boyer-Moore algorithms, and a survey questionnaire sectioned per Bloom's revised taxonomy of the cognitive domain. Based on findings, it is apparent that students have difficulties in learning the IM course as confirmed by the students' manner of learning the SQL syntax, particularly for the cognitive domain - analysing on KUs such as relational database and data modelling. In effect, the more students who failed to fully understand the prerequisite KUs, the fewer can advance to the next KUs as revealed by the 12.3% DQL error logs.