Abstract | ||
---|---|---|
In multi-turn dialogs, natural language understanding models can introduce obvious errors by being blind to contextual information. To incorporate dialog history, we present a neural architecture with Speaker-Sensitive Dual Memory Networks which encode utterances differently depending on the speaker. This addresses the different extents of information available to the system - the system knows only the surface form of user utterances while it has the exact semantics of system output. We performed experiments on real user data from Microsoft Cortana, a commercial personal assistant. The result showed a significant performance improvement over the state-of-the-art slot tagging models using contextual information. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/ASRU.2017.8268983 | 2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) |
Keywords | DocType | Volume |
speaker-sensitive dual memory networks,Microsoft Cortana,natural language understanding models,multiturn dialogs,multiturn slot tagging,incorporate dialog history,contextual information | Conference | abs/1711.10705 |
ISBN | Citations | PageRank |
978-1-5090-4789-5 | 0 | 0.34 |
References | Authors | |
9 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Young-Bum Kim | 1 | 112 | 13.60 |
Sungjin Lee | 2 | 221 | 27.44 |
Ruhi Sarikaya | 3 | 698 | 64.49 |