Abstract:
Sentiment classification is a functioning examination
area in information mining and information
disclosure with different application spaces.
Accomplishment of item advancement sites, for
example, Amazon, eBay and so forth gets influenced
by the nature of the surveys they have for their
yields. Every one of these destinations gives a path to
the analyst to form the remarks on the basis of the
items and assign a remark to it. Considering these
remarks, the analysis will be classified as best or
worst. By this, a structure can be edified that can
identify the sentiment masked in a review,
performing sentiment categorization on a gigantic
dataset. All particulars can be grouped into
primarily two classes, facts and opinions. Facts are
assertions about matter and worldly occurrences.
And opinions are individual statements that mirror
individuals' assumptions or bits of knowledge about
the entities and events. This paper shows the
performance of classification algorithms such as
Decision Tree, Bernoulli NB, Logistic, and
Perceptron using Principal Component Analysis
(PCA), applying n-gram (unigram, bigram) on the
entire feature set and computing confusion matrix
for the dataset
King Mongkut's University of Technology North Bangkok. Central Library
Address:
BANGKOK
Email:
library@kmutnb.ac.th
Created:
2021
Modified:
2024-06-11
Issued:
2024-06-10
บทความ/Article
application/pdf
BibliograpyCitation :
In IEEE Computer Society. 2021 Fourth International Conference on Computational Intelligence and Communication Technologies (CCICT 2021) (pp.116-120). Los Alamitos, CA : IEEE Computer Society