Machine Learning and Big-Data Analytics Identify Important Predictors to Students’ Science Academic Achievement Globally

This presentation aims to demonstrate how a big-data method, the Boosted-Tree, can be employed to analyze the 2015 Programme for International Student Assessment. Results indicate that out of 500 potential predictors, six variables contributed to the U.S. students' science-scores, but only two variables contributed to students’ science-scores in Asian countries.