Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Hana Chyntia Morama
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analisis Sentimen berbasis Aspek terhadap Ulasan Hotel Tentrem Yogyakarta menggunakan Algoritma Random Forest Classifier Hana Chyntia Morama; Dian Eka Ratnawati; Issa Arwani
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 4 (2022): April 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The development of tourism has increased visits in tune with the hospitality industry in Indonesia. One of the popular five-star hotels is Hotel Tentrem Yogyakarta. The large number of hotel review data makes visitors confused to make the right decision. Sentiment analysis can overcome this problem by processing review text data which was initially unstructured into information that has positive, negative or neutral values. In addition, aspect categorization is also carried out so that it is easier for visitors to find reviews according to their purpose. The aspects used in this research are room aspects, service aspects, location aspects, swimming pool aspects, and gym aspects. Hotel review data was obtained by scraping using the Webscraper.io tool on the Tripadvisor website. Classification was carried out using the Random Forest Classifier algorithm and term frequency-inverse document frequency (TF-IDF) word weighting. After analyzing the test, the aspect that is used is only the room aspect because it has a balanced proportion of sentiment compared to other aspects. The proportion of sentiment is considered important in the classification of sentiment. The test is carried out based on the parameter scenario of the number of trees and the depth of the tree. The number of trees used in this study is 300 and the depth of the tree is 10. The test results prove that the greater the number of trees and the depth of the tree, the better the prediction results. The best classification results in the room aspect is 90% for the accuracy value and the f1 score.
Co-Authors Dian Eka Ratnawati Issa Arwani