Feature extraction using CMIM for sentiment analysis
by B. Madhusudhanan; S. Chitra; S. Anbuchelian
International Journal of Advanced Intelligence Paradigms (IJAIP), Vol. 13, No. 3/4, 2019

Abstract: Recently, a lot of attention paid to the domain of sentiment analysis (SA), with experts acknowledging the scientific trials as well as possible applications of the processing of subjective language. SA is the computational analysis of opinions or sentiments conveyed in a body of text. The aim of SA is the detection of subjective data present in several sources and figure out the attitudes of the author regarding the topic. Features extraction looks after the identification of the features used for opinion mining, features selection is used for choosing best features for opinion classification, features weighting method is used for weighting features for good recommendations, reduction methods are used for optimisation of classification procedure. In the current study, the feature extraction is carried out term frequency/inverse document frequency and features selection through conditional mutual information maximisation (CMIM). Feature classification is done through LogitBoost, chi-squared automatic interaction detector (CHAID) as well as K-nearest neighbour (KNN) classifiers. The experimental results were contrasted with one another.

Online publication date: Tue, 03-Sep-2019

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Advanced Intelligence Paradigms (IJAIP):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com