E3S Web Conf.
Volume 202, 2020The 5th International Conference on Energy, Environmental and Information System (ICENIS 2020)
|Number of page(s)||10|
|Section||Information System for Economic and Business|
|Published online||10 November 2020|
Sentiment Analysis on Tokopedia Product Online Reviews Using Random Forest Method
1 Bachelor Program of Statistics, Faculty of Science and Mathematics, Diponegoro University, Indonesia Semarang – Indonesia
2 Department of Statistics, Faculty of Science and Mathematics, Diponegoro University, Semarang – Indonesia
3 Department of Statistics, Faculty of Science and Mathematics, Diponegoro University, Semarang – Indonesia
* Corresponding author: email@example.com
Tokopedia is one of the most popular e-commerce sites in Indonesia that offers consumer products from various categories. In each product section, a review feature is offered. This review feature became essential in evaluating the sellers and become one consideration for customers in making purchase consideration. Sentiment analysis of Tokopedia product reviews may provide the opportunity to look on how Tokopedia customers respond to product quality and sellers’ hospitality. In evaluating the model, the reviews were grouped as: “positive sentiment” and “negative sentiment” using the Random Forest method and 10-fold cross-validation. Data labelling was carried out automatically by calculating the sentiment score using Lexicon-Based. Visualization of the labelling results was then done using a bar graph and a word cloud on each class of sentiment in order to look up for information that is considered important and most discussed. The test results showed that the accuracy of the Random Forest Method with parameter mtry = 73 and ntree = 50 is 97.38% which leads to the conclusion that the Random Forest Method could well predict the product reviews of Tokopedia. The greater the accuracy, the better performance of the classification model.
Key words: Product Reviews / Tokopedia / Random Forest / Text mining / Word Cloud
© The Authors, published by EDP Sciences, 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.