SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER
At present, digital communication and data have become the higher use, and expressing their message through reviews and many more. The cosmetics industry has developed into a place where every business and sector competes to market and enhance its brand. Sephora, one of the biggest cosmetic industri...
Published in: | Journal of Engineering Science and Technology |
---|---|
Main Author: | |
Format: | Article |
Language: | English |
Published: |
Taylor's University
2023
|
Online Access: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184571948&partnerID=40&md5=124471313771c92cc608c7809cffa4f1 |
id |
2-s2.0-85184571948 |
---|---|
spelling |
2-s2.0-85184571948 Fadly; Kurniawan T.B.; Dewi D.A.; Zakaria M.Z.; Nazziri N.F.B.M. SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER 2023 Journal of Engineering Science and Technology 18 6 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184571948&partnerID=40&md5=124471313771c92cc608c7809cffa4f1 At present, digital communication and data have become the higher use, and expressing their message through reviews and many more. The cosmetics industry has developed into a place where every business and sector competes to market and enhance its brand. Sephora, one of the biggest cosmetic industries, has higher sales and promotion, and its website has more reviews than it could ever get. The consumer can access the reviews and view them to give their opinion. The user's opinion can be predicted to know the positive and negative. It brought us to sentiment analysis as the focus of research to see the review. The Nave Bayes classifier, which is automatically pre-processed using natural language processing, is modelled. In building the model, the process goes through data collection, such as the reviews from each product brand. Then the pre-processing is done to get the bag of words trained in the Naïve Bayes model. The data have been trained with different split ratios and the number of iterations to find the highest accuracy. Then the data will be fine-tuning to get higher accuracy results to measure the prediction. As the model goes through, the visualization shows the prediction data. As a result, the Naïve Bayes model showed 94.7% accuracy after measuring using the cross-validation technique. © 2023 Taylor's University. All rights reserved. Taylor's University 18234690 English Article |
author |
Fadly; Kurniawan T.B.; Dewi D.A.; Zakaria M.Z.; Nazziri N.F.B.M. |
spellingShingle |
Fadly; Kurniawan T.B.; Dewi D.A.; Zakaria M.Z.; Nazziri N.F.B.M. SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
author_facet |
Fadly; Kurniawan T.B.; Dewi D.A.; Zakaria M.Z.; Nazziri N.F.B.M. |
author_sort |
Fadly; Kurniawan T.B.; Dewi D.A.; Zakaria M.Z.; Nazziri N.F.B.M. |
title |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
title_short |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
title_full |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
title_fullStr |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
title_full_unstemmed |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
title_sort |
SENTIMENT ANALYSIS ON COSMETIC PRODUCT IN SEPHORA USING NAÏVE BAYES CLASSIFIER |
publishDate |
2023 |
container_title |
Journal of Engineering Science and Technology |
container_volume |
18 |
container_issue |
6 |
doi_str_mv |
|
url |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184571948&partnerID=40&md5=124471313771c92cc608c7809cffa4f1 |
description |
At present, digital communication and data have become the higher use, and expressing their message through reviews and many more. The cosmetics industry has developed into a place where every business and sector competes to market and enhance its brand. Sephora, one of the biggest cosmetic industries, has higher sales and promotion, and its website has more reviews than it could ever get. The consumer can access the reviews and view them to give their opinion. The user's opinion can be predicted to know the positive and negative. It brought us to sentiment analysis as the focus of research to see the review. The Nave Bayes classifier, which is automatically pre-processed using natural language processing, is modelled. In building the model, the process goes through data collection, such as the reviews from each product brand. Then the pre-processing is done to get the bag of words trained in the Naïve Bayes model. The data have been trained with different split ratios and the number of iterations to find the highest accuracy. Then the data will be fine-tuning to get higher accuracy results to measure the prediction. As the model goes through, the visualization shows the prediction data. As a result, the Naïve Bayes model showed 94.7% accuracy after measuring using the cross-validation technique. © 2023 Taylor's University. All rights reserved. |
publisher |
Taylor's University |
issn |
18234690 |
language |
English |
format |
Article |
accesstype |
|
record_format |
scopus |
collection |
Scopus |
_version_ |
1809677578861019136 |