Root Cause Analysis of Miscommunication Hotspots in Spoken Dialogue Systems

Spiros Georgiladakis, Georgia Athanasopoulou, Raveesh Meena, José Lopes, Arodami Chorianopoulou, Elisavet Palogiannidi, Elias Iosif, Gabriel Skantze, Alexandros Potamianos


A major challenge in Spoken Dialogue Systems (SDS) is the detection of problematic communication (hotspots), as well as the classification of these hotspots into different types (root cause analysis). In this work, we focus on two classes of root cause, namely, erroneous speech recognition vs. other (e.g., dialogue strategy). Specifically, we propose an automatic algorithm for detecting hotspots and classifying root causes in two subsequent steps. Regarding hotspot detection, various lexico-semantic features are used for capturing repetition patterns along with affective features. Lexico-semantic and repetition features are also employed for root cause analysis. Both algorithms are evaluated with respect to the Let’s Go dataset (bus information system). In terms of classification unweighted average recall, performance of 80% and 70% is achieved for hotspot detection and root cause analysis, respectively.


DOI: 10.21437/Interspeech.2016-1273

Cite as

Georgiladakis, S., Athanasopoulou, G., Meena, R., Lopes, J., Chorianopoulou, A., Palogiannidi, E., Iosif, E., Skantze, G., Potamianos, A. (2016) Root Cause Analysis of Miscommunication Hotspots in Spoken Dialogue Systems. Proc. Interspeech 2016, 1156-1160.

Bibtex
@inproceedings{Georgiladakis+2016,
author={Spiros Georgiladakis and Georgia Athanasopoulou and Raveesh Meena and José Lopes and Arodami Chorianopoulou and Elisavet Palogiannidi and Elias Iosif and Gabriel Skantze and Alexandros Potamianos},
title={Root Cause Analysis of Miscommunication Hotspots in Spoken Dialogue Systems},
year=2016,
booktitle={Interspeech 2016},
doi={10.21437/Interspeech.2016-1273},
url={http://dx.doi.org/10.21437/Interspeech.2016-1273},
pages={1156--1160}
}