DocumentCode
442046
Title
Machine learning for automatic acquisition of Chinese linguistic ontology knowledge
Author
Zheng, De-quan ; Zhao, Tie-jun ; Yu, Feng ; Sheng-Li
Author_Institution
Sch. of Comput. Sci. & Eng., Harbin Univ. of Commerce, China
Volume
6
fYear
2005
fDate
18-21 Aug. 2005
Firstpage
3728
Abstract
Due to the complexity and flexibility of natural language, automatic linguistic knowledge acquisition and its application research becomes difficult. In this paper, we present a machine learning method to automatically acquire Chinese linguistic ontology knowledge from typical corpus. This study, first, defined the description frame of Chinese linguistic ontology knowledge, and then, automatically acquired the usage of a Chinese word with its co-occurrence of context in using semantic, pragmatics, syntactic, etc from the corpus, final, the above information and their representation act as Chinese linguistic ontology knowledge bank. We completed two groups of experiments, i.e. documents similarity computing, text reordering for information retrieval. Compared with previous works, the proposed method solves the inferior precision of nature language processing.
Keywords
knowledge acquisition; learning (artificial intelligence); linguistics; natural languages; ontologies (artificial intelligence); Chinese linguistic ontology knowledge; document similarity computing; information retrieval; knowledge acquisition; machine learning; natural language processing; text reordering; Computer science; Information retrieval; Knowledge acquisition; Knowledge engineering; Learning systems; Machine learning; Natural language processing; Natural languages; Ontologies; Tagging; Machine learning; knowledge acquisition; linguistic ontology knowledge; natural language processing;
fLanguage
English
Publisher
ieee
Conference_Titel
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location
Guangzhou, China
Print_ISBN
0-7803-9091-1
Type
conf
DOI
10.1109/ICMLC.2005.1527589
Filename
1527589
Link To Document