Deciphering Meaning from Text in Deep Learning Analysis

Researchers will present findings from an educational data mining study incorporating both deep learning and text-mining of asynchronous forum posts from a K-12 virtual learning environment. Findings suggest that the inclusion of textual content improved the overall predictive accuracy in identifying at-risk students. However, there were challenges in deciphering the meaning behind the predictive text features. Results and future directions for research inquiry will be shared.