TY - GEN
T1 - Causal analysis of task completion errors in spoken music retrieval interactions
AU - Hara, Sunao
AU - Kitaoka, Norihide
AU - Takeda, Kazuya
N1 - Funding Information:
This work was supported in part by the NEDO Grant for Industrial Technology Research Program (07A12203a) and KAKENHI (23500209).
PY - 2012
Y1 - 2012
N2 - In this paper, we analyze the causes of task completion errors in spoken dialog systems, using a decision tree with N-gram features of the dialog to detect task-incomplete dialogs. The dialog for a music retrieval task is described by a sequence of tags related to user and system utterances and behaviors. The dialogs are manually classified into two classes: completed and uncompleted music retrieval tasks. Differences in tag classification performance between the two classes are discussed. We then construct decision trees which can detect if a dialog finished with the task completed or not, using information gain criterion. Decision trees using N-grams of manual tags and automatic tags achieved 74.2% and 80.4% classification accuracy, respectively, while the tree using interaction parameters achieved an accuracy rate of 65.7%. We also discuss more details of the causality of task incompletion for spoken dialog systems using such trees.
AB - In this paper, we analyze the causes of task completion errors in spoken dialog systems, using a decision tree with N-gram features of the dialog to detect task-incomplete dialogs. The dialog for a music retrieval task is described by a sequence of tags related to user and system utterances and behaviors. The dialogs are manually classified into two classes: completed and uncompleted music retrieval tasks. Differences in tag classification performance between the two classes are discussed. We then construct decision trees which can detect if a dialog finished with the task completed or not, using information gain criterion. Decision trees using N-grams of manual tags and automatic tags achieved 74.2% and 80.4% classification accuracy, respectively, while the tree using interaction parameters achieved an accuracy rate of 65.7%. We also discuss more details of the causality of task incompletion for spoken dialog systems using such trees.
KW - Interaction parameters
KW - Spoken dialog
KW - Task incompletion
UR - http://www.scopus.com/inward/record.url?scp=85037376279&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85037376279&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85037376279
T3 - Proceedings of the 8th International Conference on Language Resources and Evaluation, LREC 2012
SP - 1365
EP - 1372
BT - Proceedings of the 8th International Conference on Language Resources and Evaluation, LREC 2012
A2 - Dogan, Mehmet Ugur
A2 - Mariani, Joseph
A2 - Moreno, Asuncion
A2 - Goggi, Sara
A2 - Choukri, Khalid
A2 - Calzolari, Nicoletta
A2 - Odijk, Jan
A2 - Declerck, Thierry
A2 - Maegaard, Bente
A2 - Piperidis, Stelios
A2 - Mazo, Helene
A2 - Hamon, Olivier
PB - European Language Resources Association (ELRA)
T2 - 8th International Conference on Language Resources and Evaluation, LREC 2012
Y2 - 21 May 2012 through 27 May 2012
ER -