Konferensartikel

Predicting the knowledge of language learners is crucial for personalized interactions in any intelligent tutoring system for language learning. This study adopts a machine learning approach to the task of predicting the knowledge of single words for individual learners of English. We experiment with two machine learning models, neural networks and random forest, and with a set of learner-specific and word-specific features. Both the models are trained for all the learners together. However, since learner-specific features are used, the prediction is personalized for every learner. Both of the models achieve state-of-the-art results for the task of vocabulary prediction for English learners.

word knowledge prediction, second language vocabulary assessment, intelligent tutoring system, machine learning