In this paper we present the first work on the automated scoring of\nmindreading ability in middle childhood and early adolescence. We create\nMIND-CA, a new corpus of 11,311 question-answer pairs in English from 1,066\nchildren aged 7 to 14. We perform machine learning experiments and carry out\nextensive quantitative and qualitative evaluation. We obtain promising results,\ndemonstrating the applicability of state-of-the-art NLP solutions to a new\ndomain and task.\n
Paper
References (38)
Scroll for more · 26 remaining