Cocea, M. & Weibelzahl, S. (2010). Validation Issues in Educational Data Mining - the case of HTML-Tutor and iHelp. In: C. Romero, S. Ventura, M. Pechenizkiy, R. S.J.d. Baker (Eds.), Handbook of Educational Data Mining (pp. 377-387). Boca Raton: Chapman & Hall/CRC.

Validation is one of the key aspects in data mining and even more so in educational data mining (EMD) owing to the nature of the data. In this chapter, a brief overview of validation in the context of EDM is given and a case-study is presented. The field of the case study is related to motivational issues in general and disengagement detection in particular. There are several approaches to eliciting motivational knowledge from a learner's activity trace; in this chapter the validation of such an approach is presented and discussed. The chapter is structured as follows. In the next section an overview of validation in the context of EDM is presented. Section 2 presents the case-study, including previous work on motivation in e-Learning, details of data and methods, and results. Section 3 presents some challenges encountered and lessons learned and, finally, Section 4 concludes the paper.

author = {Mihaela Cocea and Stephan Weibelzahl},
title = {Validation {I}ssues in {E}ducational {D}ata {M}ining --- the case of {HTML}-{T}utor and i{H}elp},
booktitle = {Handbook of {E}ducational {D}ata {M}ining},
editor = {Cristobal Romero Morales and Sebastian Ventura and Silvia Rita Viola and Mykola Pechenizkiy and Ryan Baker},
publisher = {Chapman \& Hall/CRC},
place = {{B}oca {R}aton},
year = {2010},
pages = {377--387}