BioWorldParser: A suite of parsers for leveraging educational data mining techniques

There has been a dramatic expansion in both the amount of available large-scale educational databases and educational mining techniques. Educational data mining has been a fertile subject of research in recent times; further, the use of educational data mining has become popular among both researchers and practitioners. Log files generated by computer-based learning environments like Intelligent Tutoring Systems contain a wealth of information about learner behaviors that characterize academic success. There is growing interest in mining these data sources for knowledge-based discovery to reveal relevant, meaningful, and useful educational information to illuminate our understanding of learners' behaviors and outcomes. All too often however, extracting the pertinent information from the data to leverage the data mining techniques can be a major roadblock; for example, the asynchronous nature of the data logged in computer-based learning environments and data mining tools pose several challenges for mining data. We sought to mitigate this by developing a parser for the BioWorld System. In this paper, we explore the viability of a hand-coded parser by presenting BioWorldParser (a suite of scripts), which was developed to parse and retrieve data from raw log files generated by the BioWorld system, to help leverage educational data mining techniques in the context of an Intelligent Tutoring System for the medical domain.