Publication: Sentiment analysis of customer tweets related to philippine internet service providers
No Thumbnail Available
Date
2023-04
Journal Title
Journal ISSN
Volume Title
Publisher
Don Mariano Marcos Memorial State Univeristy - South La Union Campus
Abstract
The researchers conducted a sentiment analysis on customer tweets which are
related to the two chosen Philippine internet service providers to determine the company's
efficiency in delivering the best to their customers. It aimed to determine the preprocessing
techniques to be used on the extracted dataset; to train the sentiment analysis model using
machine learning algorithms; and to evaluate the performance of the models in terms of
accuracy.
The research design of this study was experimental approach. The researchers
gathered tweets as data by using RapidMiner and Twitter API Three algorithms were used
in this study: Naive Bayes with Laplace Smoothing, Support Vector Machine with Kernel
Types and K-Nearest Neighbor with K-values as parameters. Both corpora showed that
Support Vector Machine with kernel type Radial Basis Function had the highest accuracy
and was concluded to be the best algorithm and parameter. Internet Service Provider 1
has higher accuracy with 0.8581 while Internet Service Provider 2 have an accuracy score
of 0.7027 therefore the best corpus was Internet Service Provider 1.
Description
Full text.
Keywords
Citation
Acenas, J. C., Casela, J. A., Cernado, A. B., & Estacio, A. N. (2023) Sentiment analysis of customer tweets related to philippine internet service providers [Unpublished Undergraduate thesis]. Don Mariano Marcos Memorial State Univeristy - South La Union Campus, Agoo, La Union.