Predict a student's performance in high school, using Linear Regression and training multiple models.
We are using first (G1) and second (G2) period grade in a subject e.g. Mathematics, along other attributes such as student's health, free time, family relationships, school support, to predict the final (G3) grade. 33 variable attributes are available to choose from, but of course G1 and G2 should be included to make our prediction more accurate. I managed to score a 96.6% accuracy after training the model and saving the best result.
Using a dataset collected in a high school, by collecting school reports and questionnaires. The full list of attribute variables can be found on: https://archive.ics.uci.edu/ml/datasets/Student+Performance (UCI Machine Learning Repository). [P. Cortez and A. Silva. Using Data Mining to Predict Secondary School Student Performance. In A. Brito and J. Teixeira Eds., Proceedings of 5th FUture BUsiness TEChnology Conference (FUBUTEC 2008) pp. 5-12, Porto, Portugal, April, 2008, EUROSIS, ISBN 978-9077381-39-7. ]