A Tutorial Markov Analysis of Effective Human Tutorial Sessions
Nabin Maharjan, Vasile Rus · 2018
This paper investigates what differentiates effective tutorial sessions from less effective sessions.Towards this end, we characterize and explore human tutors' actions in tutorial dialogue sessions by mapping the tutor-tutee interactions, which are streams of dialogue utterances, into streams of actions, based on the language-as-action theory.Next, we use human expert judgment measures, evidence of learning (EL) and evidence of soundness (ES), to identify effective and ineffective sessions.We perform sub-sequence pattern mining to identify sub-sequences of dialogue modes that discriminate good sessions from bad sessions.We finally use the results of subsequence analysis method to generate a tutorial Markov process for effective tutorial sessions.