Quantitative Analysis of Student Data Mining

Abstract

Educational Data Mining (EDM) in the research field will constitute an application in major techniques like Data Mining, Machine Learning and Statistical Techniques in the education and organization sector. It aims at defining better mechanisms to analyze student performance by use of sophisticated predictive techniques. The insights from the analysis based on their previous performances can be used for future performance prediction, counselling students for university enrollment and to help them select electives for their undergraduate courses. This also helps students decide on their career path and the colleges to monitor student performance at any given time, so that it can be used as a record for their placements. To gain insights from the data one can choose their own metrics. The major challenge in this is to capture and clean the data and also to find an appropriate technique to carry on with the analysis. This paper introduces best suited methods to capture data by choosing the right metrics and performance indicators. In EDM, a right metric is the one that is unbiased and which considers all the aspects whether academic or non-academic. Through this paper we also intend to help analysts decide on choosing the right approach, techniques and algorithms to start with EDM.