Towards emotion prediction in spoken tutoring dialogues

Published 1 January 2003Open access

Diane Litman, Kate Forbes, Scott Silliman

Citations28

TL;DR

Preliminary machine learning experiments involving transcription, emotion annotation and automatic feature extraction from the human-human spoken tutoring corpus indicate that the spoken tutor system the developing can be enhanced to automatically predict and adapt to student emotional states.

Abstract

Human tutors detect and respond to student emotional states, but current machine tutors do not. Our preliminary machine learning experiments involving transcription, emotion annotation and automatic feature extraction from our human-human spoken tutoring corpus indicate that the spoken tutoring system we are developing can be enhanced to automatically predict and adapt to student emotional states.

Keywords

Computer Science

Prosody-based automatic detection of annoyance and frustration in human-computer dialog

388 Citations2002Jeremy Ang, Rajdip Dhillon +3 more

Learning trees and rules with set-valued features

244 Citations1996William W. Cohen

It is argued that many decision tree and rule learning algorithms can be easily extended to set-valued features, and it is shown by example that many real-world learning problems can be efficiently and naturally represented with set- valued features.

Lecture notes in computer scienceThe Architecture of Why2-Atlas: A Coach for Qualitative Physics Essay Writing

232 Citations2002Kurt VanLehn, Pamela Jordan +10 more

The Why2-Atlas system teaches qualitative physics by having students write paragraph-long explanations of simple mechanical phenomena, and the tutor uses deep syntactic analysis and abductive theorem proving to convert the student's essay to a proof.

User Modeling and User-Adapted InteractionDesigning and Evaluating an Adaptive Spoken Dialogue System

159 Citations2002Diane Litman, Shimei Pan

This paper presents the design and evaluation of an adaptive version of TOOT, a spoken dialogue system for retrieving online train schedules that constructs a user model representing whether the user is having speech recognition problems as a particular dialogue progresses.

Experimentally augmenting an intelligent tutoring system with human-supplied capabilities: adding human-provided emotional scaffolding to an automated reading tutor that listens

70 Citations2003Greg Aist, Barry Kort +3 more

Human-provided emotional scaffolding added to the automated Reading Tutor resulted in increased student persistence, compared to the reading Tutor alone, and the general technique of adding human-supplied capabilities to an existing intelligent tutoring system should prove useful for studying other ITSs too.

Artificial intelligenceThe Recognition of Emotion

40 Citations2000Anton Batliner, Richard Huber +4 more

A module is described that combines several knowledge sources within an integrated classification of trouble in communication within Verbmobil for the modelling of emotional user behavior.

Lecture notes in computer scienceAdding Human-Provided Emotional Scaffolding to an Automated Reading Tutor That Listens Increases Student Persistence

26 Citations2002Gregory Aist, Barry Kort +3 more

This poster presents the first statistically reliable empirical evidence from a controlled study for the effect of human-provided emotional scaffolding on student persistence in an intelligent tutoring system.

Predicting user reactions to system error

23 Citations2001Diane Litman, Julia Hirschberg +1 more

Comparisons of features of aware sites in a train timetable spoken dialogue corpus reveal significant prosodic differences between such turns, compared with turns that 'correct' speech recognition errors as well as with 'normal' turns that are neither aware sites nor corrections.

View PDF