DocumentCode
3131639
Title
Transaction Dialogs Based on Short Messages Understanding
Author
Ye, Shiren ; Liao, Ding-an ; Zhou, Ye
Author_Institution
Sch. of Inf. Technol., Univ. of Changzhou, Changzhou, China
fYear
2011
fDate
8-9 Oct. 2011
Firstpage
117
Lastpage
120
Abstract
The paper proposed the crucial technique in transaction service system via Short Messages (SMs), which is a novel kind of dialog based on SMs understanding. After determining the domain of a new dialog using SPN, the service agent SMartS extracts information from SMs that are parsed by SM parser, part-of-speech tagger, name entities recognizer and semantic parser in turn, and then replies the prompt which could direct the procedure of dialog. An induction algorithm is used to discover automatically most domain-specific parser rules from the corpus which contains original SMs and the predefined case frame which depicts detailed semantic groups. Finally, a real domain application was provided.
Keywords
grammars; information retrieval; interactive systems; software agents; speech processing; SM parser; SPN; domain-specific parser rules; entities recognizer; information extraction; part-of-speech tagger; semantic parser; service agent SMartS; short messages; transaction dialogs; transaction service system; Accuracy; Business; Data mining; Satellite broadcasting; Semantics; Syntactics; Training; dialog; message understanding;
fLanguage
English
Publisher
ieee
Conference_Titel
Knowledge Acquisition and Modeling (KAM), 2011 Fourth International Symposium on
Conference_Location
Sanya
Print_ISBN
978-1-4577-1788-8
Type
conf
DOI
10.1109/KAM.2011.39
Filename
6137593
Link To Document