MOOC Optimization Dataset: Learner Feedback and AI-Based Pareto Solutions

his dataset accompanies the study “AI-Based Multi-Objective Optimization for MOOC Updates Using Learner Feedback”. It provides the full set of inputs, outputs, and code used to generate the results reported in the paper.

The dataset includes:

• candidate_updates.csv — Input dataset derived from 437 learner feedback responses, listing potential course updates with associated learner value, effort units, and category.

• pareto_B8.csv, pareto_B10.csv, pareto_B12.csv, pareto_B14.csv, pareto_B16.csv — Complete Pareto-optimal solution sets for five budget levels. Each file reports learner value, effort, diversity index, and selected updates.

• optimization_nsga2.py — Python implementation of the NSGA-II optimization model used to generate Pareto-optimal solutions.

• README.txt — Documentation with instructions for running the code and reproducing results.

This resource allows full reproducibility of the optimization study and may be reused for further research on MOOCs, optimization, and educational technology.