Superlative model using word cloud for short answers evaluation in eLearning

Abstract

Teachers are thrown open to abundance of free text answers which are very daunting to read and evaluate. Automatic assessments of open ended answers have been attempted in the past but none guarantees 100 % accuracy. In order to deal with the overload involved in this manual evaluation, a new tool becomes necessary. The unique superlative model discussed in this paper aims at providing improved accuracy by constructing word clouds. The model uses appropriate semantics with a visual appeal to partially automate free text evaluation. The model was applied at a K-12 school setup where the average human agreement rate was found to be 98 % and the accuracy score deviation from the mean was 2.82. This tool can be cast-off at any level starting from K-12 to higher education to evolve the way we view and evaluate answers.

from # All Medicine by Alexandros G. Sfakianakis via Alexandros G.Sfakianakis on Inoreader http://ift.tt/2wdQpvl
via IFTTT