Answer to "What is educational data mining?"


Topic Editor: Dr. Martin Pusic, New York University


Educational data mining (EDM) refers to the use of computer algorhythmic approaches to uncovering patterns in the ever larger collections of digital data collected about learners.  These come in three flavors:  prediction, structure discovery and relationship discovery.  In prediction, the goal is to develop a model that can infer a single predicted variable from all the other variables.  The techniques can range from regression models to deep learning by artificial neural networks.  For structure discovery, the goal is to uncover the underlying structure in the data.  Factor and cluster analysis are dominant techniques thought social network analysis is an increasingly prevalent.  Finally, in relationship mining, the goal is to determine valid relationships between elements of the data.  These can be at the association or causal level depending on the type and magnitude of data available.  Sequential pattern mining uses temporal information to map paths within an educational space. 


  1. Baker RS, Inventado PS. Educational data mining and learning analytics. In Learning analytics 2014 (pp. 61-75). Springer New York

  2. Baker, R. S., & Siemens, G. (2014). Educational data mining and learning analytics. In R. Sawyer (Ed), The Cambridge handbook of the learning sciences (pp. 253–272). Cambridge University Press.

© 2019 AMEE