TY - JOUR
T1 - Automatic prediction of frustration
AU - Kapoor, Ashish
AU - Burleson, Winslow
AU - Picard, Rosalind W.
N1 - Funding Information:
We would like to thank Selene Mota for her software for the chair posture pattern analysis, and also many past members of the Affective Computing group at MIT for their work developing sensors and software, especially Carson Reynolds and Marc Strauss. We are indebted to Ken Perlin and John Lippincott for their development of the agent, and to John Rebula for help with system development, data collection and analysis. We would like to thank John D’Auria, Principal of Wellesley Middle School, and their students and staff for being so accommodating of this experiment. We would like to thank Steelcase for loaning us the posture sensor. This material is based upon work supported in part by the Media Lab Things That Think consortium, Deutsche Telekom, by funds from NASA via Ahmed Noor, and by the National Science Foundation under ITR Grant No. 0325428. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Science Foundation or the official policies, either expressed or implied, of the sponsors or of the United States Government.
PY - 2007/8
Y1 - 2007/8
N2 - Predicting when a person might be frustrated can provide an intelligent system with important information about when to initiate interaction. For example, an automated Learning Companion or Intelligent Tutoring System might use this information to intervene, providing support to the learner who is likely to otherwise quit, while leaving engaged learners free to discover things without interruption. This paper presents the first automated method that assesses, using multiple channels of affect-related information, whether a learner is about to click on a button saying "I'm frustrated." The new method was tested on data gathered from 24 participants using an automated Learning Companion. Their indication of frustration was automatically predicted from the collected data with 79% accuracy (chance = 58 %). The new assessment method is based on Gaussian process classification and Bayesian inference. Its performance suggests that non-verbal channels carrying affective cues can help provide important information to a system for formulating a more intelligent response.
AB - Predicting when a person might be frustrated can provide an intelligent system with important information about when to initiate interaction. For example, an automated Learning Companion or Intelligent Tutoring System might use this information to intervene, providing support to the learner who is likely to otherwise quit, while leaving engaged learners free to discover things without interruption. This paper presents the first automated method that assesses, using multiple channels of affect-related information, whether a learner is about to click on a button saying "I'm frustrated." The new method was tested on data gathered from 24 participants using an automated Learning Companion. Their indication of frustration was automatically predicted from the collected data with 79% accuracy (chance = 58 %). The new assessment method is based on Gaussian process classification and Bayesian inference. Its performance suggests that non-verbal channels carrying affective cues can help provide important information to a system for formulating a more intelligent response.
KW - Affect recognition
KW - Affective Learning Companion
KW - Intelligent Tutoring System
KW - Learner state assessment
UR - http://www.scopus.com/inward/record.url?scp=34249276773&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34249276773&partnerID=8YFLogxK
U2 - 10.1016/j.ijhcs.2007.02.003
DO - 10.1016/j.ijhcs.2007.02.003
M3 - Article
AN - SCOPUS:34249276773
SN - 1071-5819
VL - 65
SP - 724
EP - 736
JO - International Journal of Human Computer Studies
JF - International Journal of Human Computer Studies
IS - 8
ER -