Application of Data Mining Techniques in Designing Knowledge Base on Student Competency at the Post Graduate level

The most valuable asset of any organization is its Knowledge and the ability to use this knowledge appropriately to enhance the output.. For an Educational Institution, the core strength would be the indepth knowledge of the predictors of a student’s performance. Many independent variables such as Gender, Educational background, Work experience might influence a candidate’s performance in the Post Graduate level. Information generated about the candidate can be converted into fruitful knowledge about future trends. A very promising tool to attain this objective is the use of appropriate Data Mining technique. The purpose of this study was to identify valid predictors and measures of the academic performance of MBA students. With the use of an effective data mining tool, an institution can determine relationships among "internal" factors such as Gender, Educational background, Work experience, and "external" factor such as student performance in their Post graduation. Association rules identify collections of data attributes that are statistically related in the underlying data. Given a database of transactions, a minimal confidence threshold and a minimal support threshold, find all association rules whose confidence and support are above the corresponding thresholds. This paper aims at using Association Rule Mining technique for classifying students based on various parameters such as their school board exam performance, specialization in the their under graduation level, Gender and their previous work experience if any. This is done by comparing their performance in the internal and external levels in their Post Graduate education.