DocumentCode :
1581136
Title :
Sentiment analysis approach to adapt a shallow parsing based sentiment lexicon
Author :
Desai, Jayraj M. ; Andhariya, Swapnil R.
Author_Institution :
Comput. Sci. & Eng. Dept., Parul Inst. of Technol., Vadodara, India
fYear :
2015
Firstpage :
1
Lastpage :
4
Abstract :
With the rapid growing of IT development and e-commerce web sites, increasing trends in people to posting online reviews. Sentiment lexicons has offend used to analyzing the large volume of online review data available and gain useful knowledge from it. Most of the sentiment lexicon are aspect base, uses dependence parsing for extracting the word which are not be able to classify the sentimental word so accurately. Try to propose a method which combines sentiment lexicon and shallow parsing. Which determine aspect and domain base sentiment analysis and then assign polarity to a lexicon. Main merits of proposed methods is that it highly accurate and automatically generating structured to avoiding the cost of manually labelling data. The shallow parsing used to analyses sentence and get the constituents words. It will not considering the internal structure of constituent word, nor specifying their value in sentence. Then using polarity of words positive or negative evolution of the product conclude.
Keywords :
data mining; grammars; pattern classification; IT development; aspect base sentiment analysis; constituent word internal structure; domain base sentiment analysis; e-commerce Web sites; online review data volume analysis; sentiment analysis approach; sentimental word classification; shallow parsing based sentiment lexicon; word extraction; Conferences; Feature extraction; Labeling; Sentiment analysis; Web mining; Aspect; Opinion Mining; Sentiment Analysis; Sentiment Lexicon; Shallow parsing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Innovations in Information, Embedded and Communication Systems (ICIIECS), 2015 International Conference on
Conference_Location :
Coimbatore
Print_ISBN :
978-1-4799-6817-6
Type :
conf
DOI :
10.1109/ICIIECS.2015.7193160
Filename :
7193160
Link To Document :
بازگشت