Detecting Certainness in Spoken Tutorial Dialogues

Detecting Certainness in Spoken Tutorial DialoguesJackson Liscombe,JuliaHirschberg,Jennifer J. VendittiSpoken Language Processing GroupDepartment of Computer ScienceColumbiaUniversityNew York City, NY, USAjaxin,julia,jjv @cs.columbia.eduAbstract ... 9.7 min. into dialogue ...TUTOR: So when you apply a force what is the resultWhat role does affect play in spoken tutorial systems and isof application of force on a body?it automatically detectable? We investigated the classificationSTUDENT: The force is transferred to the container.of student certainness in a corpus collected for ITSPOKE, a(UNCERTAIN)speech-enabled Intelligent Tutorial System (ITS). Our studysuggeststhattutorsrespondtoindicationsofstudentuncertainty TUTOR: No. Force does not get transferred todifferently from student certainty. Results of machine learning anything.experiments indicate that acoustic-prosodic features can distin- STUDENT: Uh-huh. (NEUTRAL)guish student certainness from other student states. A combi-TUTOR: Force is exerted and what does the forcenation of acoustic-prosodic features extracted at two levels ofproduce?intonational analysis — breath groups and turns — achievesSTUDENT: Movement of the container. (UNCERTAIN)76.42% classification accuracy, a 15.8% relative improvementoverbaseline performance. Ourresultssuggest thatstudent cer- TUTOR: No. acceleration.tainnesscanbeautomaticallydetectedandutilizedtocreatebet-STUDENT: Acceleration. OK. (CERTAIN)ter spoke dialog ...