Pranav Manjunath

Summary

Project has been published in Algorithms for Intelligent Systems - Cybernetics, Cognition and Machine Learning Applications (Spinger Publication) and was presented at the ICCCMLA 2019 - Goa, India.

Deciding the right graduate level university is a tedious process for any undergraduate student, due to lack of transparency in the admission process. Data mining methods such as Association Rule Mining, specifically Apriori methods, and Decision Tree Classification are two Data Mining techniques that we have employed to evaluate the graduate admission requirements in the United States of America. Apriori algorithm can elucidate key determinants for graduate admissions based on the historical data. Predictive model built with decision trees accurately determines the outcome for college admission based on the information provided by the student. Together, these two methods can provide relatively accurate information to students to make informed decisions on the choice of universities. In this study, we have used these methods successfully on the data collected from a survey by students following their admission to specific universities. We have clearly delineated specific requirements for admissions to top-, mid- and lower ranking universities. The data indicates the importance of key attributes such as research, analytical skills and language skills for entry to top notch universities.