Feature Extraction Performance on Classified Methods for Text Sentiment Analysis

P. Prihatini, K. Indah, G. Sukerti, I Indrayana, I Sudiartha

2021

Abstract

The travellers reviews for hotel services displayed by the online travel agent application have drawbacks because the text must be read one by one from all the existing reviews, and then the reader must conclude his own impression of the hotel. Through the Sentiment Analysis technique, each review text can be classified as a positive or negative impression automatically, where the impression can be taken into consideration by tourist in choosing hotel and for hotel manager in improving services improvement. To produce an appropriate classification, sentiment analysis relies on the feature extraction method and the classification technique used. This paper evaluates the performance of Term Frequency Inverse Document Frequency as feature extraction method in the five classification techniques: Support Vector Machine, Decision Tree, Random Forest, KNearest Neighbors, and Multi-Layer Perceptron, to find out which classification technique are better implemented to the dataset so it can produce the right impression. The evaluation results show that the performance of Term Frequency Inverse Document Frequency is best implemented in Support Vector Machine with a Precision value of 0.93, Recall of 1.00, and P-Score of 0.96.

Paper Citation

in Harvard Style

Prihatini P., Indah K., Sukerti G., Indrayana I. and Sudiartha I. (2021). Feature Extraction Performance on Classified Methods for Text Sentiment Analysis. In Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES, ISBN 978-989-758-615-6, pages 1235-1243. DOI: 10.5220/0010962900003260

in Bibtex Style

@conference{icast-es21,
author={P. Prihatini and K. Indah and G. Sukerti and I Indrayana and I Sudiartha},
title={Feature Extraction Performance on Classified Methods for Text Sentiment Analysis},
booktitle={Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES,},
year={2021},
pages={1235-1243},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010962900003260},
isbn={978-989-758-615-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science - Volume 1: iCAST-ES,
TI - Feature Extraction Performance on Classified Methods for Text Sentiment Analysis
SN - 978-989-758-615-6
AU - Prihatini P.
AU - Indah K.
AU - Sukerti G.
AU - Indrayana I.
AU - Sudiartha I.
PY - 2021
SP - 1235
EP - 1243
DO - 10.5220/0010962900003260