Abstract | ||
---|---|---|
Multi-domain dialog systems often encounter user requests for out-of-domain (OOD) service. This paper focuses on detecting these requests. The proposed OOD detection method is included in a multi-domain detection component naturally. This component consists of multiple in-domain verifiers: an in-domain verifier accepts a user utterance when it belongs to the domain and rejects the utterance otherwise. To detect OOD requests without using an actual OOD corpus, the in-domain verifiers exploit the occurrence of out-of-vocabulary words. In experiments, the proposed OOD detection method was more accurate than three baseline methods. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/BIGCOMP.2014.6741429 | BigComp |
Keywords | Field | DocType |
binary classification,out-of-domain detection,multidomain dialog systems,vocabulary,user utterance,ood service,multidomain detection component,multi-domain detection,out-of-vocabulary words,interactive systems,ood detection method,multiple in-domain verifiers,training data | Training set,Dialog box,Binary classification,Computer science,Utterance,Speech recognition,Exploit,Artificial intelligence,Natural language processing,Out of vocabulary,Vocabulary | Conference |
ISSN | Citations | PageRank |
2375-933X | 0 | 0.34 |
References | Authors | |
9 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Seonghan Ryu | 1 | 23 | 5.73 |
Donghyeon Lee | 2 | 82 | 8.06 |
Gary Geunbae Lee | 3 | 932 | 93.23 |
Kyungduk Kim | 4 | 154 | 12.10 |
Hyungjong Noh | 5 | 60 | 7.75 |