EssayAid: towards a semi-automatic system for assessing student texts
Tuomo Kakkonen, Erkki Sutinen · International Journal of Continuing Engineering Education and Life-Long Learning · 2011
This article introduces a model for the assessment of free text that combines both computerised and human models of assessment. This semi-automatic assessment model represents a new generation of assessment models, which offer the benefits of fully automatic grading and feedback together with the advantages of a human assessor. A model such as this must take into account the various linguistic, technical and pedagogical issues that affect the assessment of texts. In this work, the model is approached from three distinct perspectives. In order to be pedagogically sound, the model has to be based on the findings of research exploring written compositions and writing education. A model for assessing writing has to be firmly grounded in linguistics. A system based on this model can only be made functional if the technological requirements are understood and fulfilled.