Machine Learning Algorithms to Predict Potential Dropout in High-Schools

EasyChair Preprint 3920

Abstract

The availability of an immense amount of data has enabled the development of data science and consequently, its application in Education Institutions. Educational data mining enables the educator/teacher to monitor student requirement and provide the necessary response and counselling. Building a database/Repository of features like academic performance, available facilities etc. to help in the process of introducing automation in the education sector. With the help of predictive analysis, educational institutions can analyze which students may need more attention, thus modifying teaching methods to achieve the goal of 0% dropout rate. Classification algorithms like logistic regression, decision trees and K-Nearest Neighbours to predict whether a student will drop out or continue his/her education. The primary objective of this research paper is to analyze the extent to which DS/AI solutions can be applied to the education system so that students can perform better and the quality of overall education is improved.

Keyphrases: Classification, Decision Tree, Dropout detection, Educational Data Mining, KNN, Predictive Analysis, ROC curve, logistic regression

@booklet{EasyChair:3920,
  author    = {Vaibhav Singh Makhloga and Kartikay Raheja and Rishabh Jain and Orijit Bhattacharya},
  title     = {Machine Learning Algorithms to Predict Potential Dropout in High-Schools},
  howpublished = {EasyChair Preprint 3920},
  year      = {EasyChair, 2020}}