BIT students’ performance analysis with KNIME Analytics Platform

Abstract

The ability to analyse students’ behaviours and make discussions, recommendation or future predictions is the core idea of Educational Data Mining. Universities are more and more interested to implement this data mining echnique for better students’ performance and good school reputation. In this thesis, we are analysing students’ behaviours of Business Information Technology degree program in Haaga-Helia University of Applied Sciences to assess current trends for future recommendations and improvements. There is provided valuable information for understanding the environment we are working on such as statistical analysis, machine learning, data mining and educational data mining. Two iterations will be implemented for reaching the desired level of results in this analysis. The first iteration will provide a brief description on the KNIME Analytics Platform project development with the data collected from the questionnaire. On the second iteration, the results provided by the first iteration will be interpreted into valuable conclusions, observations as well as suggestions on student’s performance. Interpretation is done based the Niemivirta study of 8 scale factors of student performance. This thesis is target to Haaga-Helia University of Applied Sciences’ pedagogical and administrative staff, academic advisors and students. However, psychologists, data analysers and anyone else who is interested in Educational Data Mining and students’ behaviours can find this thesis useful. To summarize the value of this thesis is to give a big picture of the current level of student performance for the given dataset and analyse how this performance is affected from each scale factor mentioned in the Niemivirta study.