Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Sinkron : Jurnal dan Penelitian Teknik Informatika
Tiara Syavitri Rambe
Universitas Labuhanbatu
Computer Science & IT

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

Found 1 Documents
Search

 1 
Sentiment Analysis of Beauty Product Applications using the Naïve Bayes Method Tiara Syavitri Rambe; Mila Nirmala Sari Hasibuan; Muhammad Halmi Dar
Sinkron : jurnal dan penelitian teknik informatika Vol. 8 No. 2 (2023): Research Article, Volume 8 Issue 2 April, 2023
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v8i2.12303

Abstract

The number of beauty products that appear on the market makes every producer compete in attracting consumers. One of the facilities provided by manufacturers to make it easier for consumers to shop is an online shopping application that can be accessed via gadgets. Where the feature of the application is the availability of user review services User reviews are often used as a recommendation for the product to be purchased. The more positive the reviews that appear, the greater the consumer's confidence to buy the product; conversely, the more negative the reviews that appear, the more reluctant consumers are to buy. This study aims to find out how much accuracy the Naïve Bayes algorithm has in conducting sentiment analysis on user reviews of beauty product applications with different combinations of training and test data. Furthermore, it is also important to know the frequency of words that often appear in the review. The sentiment class used is divided into three, namely, positive, negative, and neutral. This research method includes a number of stages, namely: data collection, data labeling, text pre-processing, data visualization, TF-IDF, sentiment analysis, etc., until the results are obtained. This research has produced the highest accuracy rate of 90.08% in the Naïve Bayes algorithm, with a composition of 90% training data and 10% test data. While the word that often appears in user reviews is "application," with a frequency of 446 occurrences, it is followed by the word "product," 444 times, and the word "price," 312 times. The greater the amount of training data used, the higher the level of accuracy resulting from the Naïve Bayes algorithm. Meanwhile, the greater the amount of test data used, the lower the resulting accuracy value.
Co-Authors Mila Nirmala Sari Hasibuan Muhammad Halmi Dar