Master's Project : Prediction and Personalization in Massive Open Online Courses (MOOCs) environment

Guide: Prof. Ramkumar Rajendran, Education Technology, IIT Bombay

Objective:

Develop learner model in MOOC to provide personalized recommendations to avoid dropouts based on their interaction with system.

Methodology:

o Metadata tagging, cleaning, coding and analysis of collected data of 24K students’ profiles and performance data in a data science and programming course provided by NPTEL (National Programme on Technology Enhanced Learning) at IIT Madras.

o Python libraries used during the project: NumPy, Pandas, Matplotlib, Scikit-Learn, seaborn, SciPy, Statsmodels

o Development and Evaluation of regression models to predict the students' performance in qualifier examination of course.

o Impact analysis of conclusions found at the end of project and personalization of Massive Open Online Courses environment

Future Work:

o Combination of Prediction models with explanatory models to support the final step after predictions: interventions.