PhD Defense - Everaldo Aguiar
|Start:||7/10/2015 at 1:00PM|
|End:||7/10/2015 at 4:55PM|
|Location:||384 Nieuwland Hall of Science|
Faculty and students are welcome to attend the presentation portion of the defense. Light refreshments and snacks will be served.
July 10, 2015 1:00 pm 384 Nieuwland Hall of Science
Advisor(s): Dr. Nitesh Chawla and Dr. Jay Brockman
Dr. Alex Ambrose Dr. Kevin Bowyer Dr. Sidney D'Mello
“Identifying Students At Risk And Beyond: A Machine Learning Approach”
A common denominator across different cultures and societies is that people, through a variety of unique ways, are on a continuous quest to move ahead in life. While it is a fact that success can be reached from many angles, it is also well known that a person's level of education can often be a strong indicator of their achievements, professional or otherwise. Nonetheless, persisting through the education ladder can be a major challenge for some students. With that in mind, educators have always sought ways to identify students that may be struggling, as they would very likely benefit from individual attention. Generally speaking, academic attrition has caused dropout rates to become a major concern for both secondary and post-secondary institutions, which have been dedicating an increasingly larger amount of effort to the early identification of such students. In this work, we explore the application of various machine learning techniques to the problem of predicting which students may be at at risk of some adverse outcome. We compare the performance of such methods to that of more traditional approaches, discuss what specific student-level data features can be used to measure student risk, and delineate strategies that can be followed by academic institutions that wish to deploy and evaluate these predictive models.