Paper: Exploring Effective Dialogue Act Sequences in One-on-one Computer Science Tutoring Dialogues

ACL ID W11-1408
Title Exploring Effective Dialogue Act Sequences in One-on-one Computer Science Tutoring Dialogues
Venue Innovative Use of NLP for Building Educational Applications
Session
Year 2011
Authors

We present an empirical study of one-on- one human tutoring dialogues in the domain of Computer Science data structures. We are interested in discovering effective tutor- ing strategies, that we frame as discovering which Dialogue Act (DA) sequences corre- late with learning. We employ multiple lin- ear regression, to discover the strongest mod- els that explain why students learn during one-on-one tutoring. Importantly, we define “flexible” DA sequence, in which extraneous DAs can easily be discounted. Our experi- ments reveal several cognitively plausible DA sequences which significantly correlate with learning outcomes.