DocumentCode
3301263
Title
Sentiment and sentimental agent identification based on sentimental sentence dictionary
Author
Liu, Dong ; Quan, Changqin ; Ren, Fuji ; Chen, Peng
Author_Institution
Res. Center of Sci. & Technol., Beijing Univ. of Posts & Telecommun., Beijing
fYear
2008
fDate
19-22 Oct. 2008
Firstpage
1
Lastpage
5
Abstract
This paper presents a method of sentiment and sentimental agent identification based on Chinese sentimental sentence dictionary. Our method can identify eight kinds of sentiment (including joy, sorrow, love, disgust, surprise, anxiety, anger and hate), and the main sentimental agent. Sentimental sentence dictionary is composed by some sentimental sentence patterns. And the sentiment of a candidate sentence is identified by calculating the consistency with the sentimental sentence patterns. Especially, using the sentimental sentence dictionary, we can get rid of the sentences without sentiment, but have sentimental words. According to sentiment words and expressions, the sentences which are not appeared in the sentimental sentence dictionary also can be identified. The experiments show that we can get precision of 84% for sentiment identification, and precision of 69% for sentimental agent identification.
Keywords
data mining; dictionaries; natural language processing; text analysis; word processing; Chinese sentimental sentence dictionary; sentiment agent identification; sentimental agent identification; sentimental sentence patterns; Dictionaries; Information science; Intelligent agent; Intelligent systems; Laboratories; Man machine systems; Publishing; Robots; Systems engineering and theory; Telecommunication computing; Sentiment identification; sentimental agent identification; sentimental sentence dictionary;
fLanguage
English
Publisher
ieee
Conference_Titel
Natural Language Processing and Knowledge Engineering, 2008. NLP-KE '08. International Conference on
Conference_Location
Beijing
Print_ISBN
978-1-4244-4515-8
Electronic_ISBN
978-1-4244-2780-2
Type
conf
DOI
10.1109/NLPKE.2008.4906802
Filename
4906802
Link To Document