DocumentCode :
398197
Title :
Agent-assisted document composition in specific domains
Author :
Chang, Kai-Liang ; Jiau, Hewijin Christine
Author_Institution :
Dept. of Electr. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan
fYear :
2003
fDate :
30 Sept.-4 Oct. 2003
Firstpage :
444
Lastpage :
449
Abstract :
When composing a document, an author usually needs to collect the materials relevant to the composition contexts and devise a fluent logical structure to arrange the composition materials acquired. In our system, these tasks, material preparation, logical structure devising, and material deployment, are deployed to the corresponding agents. The agents acquiring composition materials from various information sources cooperate to simplify the material preparation task. These agents can consult the knowledge facilities capturing the structures and regularities in specific information sources to improve their information extraction task. There are agents devising logical structures for authors´ compositions by extracting a representative template from documents in a similar writing perspective. These agents not only consult the knowledge facilities describing the structures of the documents they analyze, but also feed useful information to augment the knowledge facilities. There are also agents deploying the composition materials acquired to the template constructed. They can learn from their interactions with authors and feed back useful information to the knowledge facilities embodying their strategies of material deployment.
Keywords :
document handling; information analysis; knowledge acquisition; software agents; agent-assisted document composition; information extraction task; information source; knowledge facility; logical structure devising agent; material deployment; material preparation task; Composite materials; Data mining; Feeds; Humans; Information analysis; Knowledge acquisition; Materials preparation; Ontologies; Performance analysis; Writing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Integration of Knowledge Intensive Multi-Agent Systems, 2003. International Conference on
Print_ISBN :
0-7803-7958-6
Type :
conf
DOI :
10.1109/KIMAS.2003.1245084
Filename :
1245084
Link To Document :
بازگشت