Identifying Disengaged Learners using Educational Data Mining Techniques

摘要

On-line learning is a new emerging technology which is dynamic and potentially enriching forms of learning but attrition remains a serious problem. Motivation towards learning is affected by the learner's self-efficacy, locus of control, goal orientation and perceived task difficulty. In a traditional classroom environment, tutors infer learners' levels of motivation from several cues, including speech, behaviour, attendance, body language or feedback, and offer interventional strategies aimed at increasing motivation. Similarly, Online learning system also needs an ability to recognize when the learner is becoming de-motivated and to intervene with effective motivational strategies. Being able to automatically detect disengaged learners would offer the opportunity to make online learning more efficient, enabling tutors and systems to target disengaged learners, to reengage them, and thus, to reduce attrition. Analysing data from log-file is an efficient method for automatic analysis, whereas it has certain level of fuzziness in order to retrieve desired information in robust fashion. Generally, the log files have more information about the learner's attitude and it does not include the results of the assessments they attend. The main purpose of extracting the disengaged students is to prevent from disengagement. To do so, the log file analysis alone could not have enough data to support our aim. Thus integration of log file information with database gives meaningful insights. Thus our proposed methodology predicts the disengagement based on the learner's attitude and Assessment performance in connection with time spent on learning based on the regional index.