Analisis Sentimen Pengguna Youtube terhadap Polemik Pelarangan Tiktok Shop menggunakan Algoritma Naive Bayes

Muhamad farhan Tholhah hidayat, Martanto Martanto, Umi Hayati

Abstract


Youtube and TikTok are creative platforms for creating videos and interacting with users. In addition to its function as a creative platform, TikTok Shop has recently emerged as a new breakthrough in the world of e-commerce because it can combine social media and e-commerce in one platform. TikTok Shop has become controversial as it disrupts micro, small, and medium-sized enterprises (MSMEs). Due to this controversy, the Indonesian government, through the Ministry of Home Affairs under the instruction of the President of Indonesia, has officially prohibited the use of TikTok as an e-commerce platform and limited it to only being a social media or social commerce application, leading to controversy turning into polemics. This has elicited various reactions from TikTok users, MSMEs, the general public, sellers, and TikTok Shop customers. Therefore, a method is needed to classify reviews automatically by conducting sentiment analysis. In this study, 4403 comment data from one CNN YouTube content titled 'TikTok Shop Banned? Ministry of Cooperatives and SMEs: If Not Regulated, Our MSMEs Could Collapse' were collected. This research applied the naïve Bayes algorithm with a qualitative and quantitative integration method and used the Knowledge Discovery in Databases (KDD) approach and confusion matrix evaluation. The data were divided into training and test sets using four schemes: first scheme 90-10, second scheme 80-20, third scheme 70-30, and fourth scheme 60-40. After evaluating the third scheme with a 70-30% data split, it achieved the best accuracy with a 94% accuracy rate of the test data in the naïve Bayes confusion matrix, which is the percentage of successfully predicted data. Furthermore, the Recall value was 96%, Precision 98%, and F1-Score 96%. This indicates that the model has a high level of accuracy for all training and test data.


Keywords


TikTokShop, YouTube, confusion matrix, KDD, Naive Bayes

Full Text:

PDF

References


Ardiansyah, D. et al. (2023) ‘Analisis Sentimen Review Pada Aplikasi Media Sosial Tiktok Menggunakan Algoritma K-Nn Dan Svm Berbasis Pso’, Jurnal Informatika Kaputama (JIK), 7(2), pp. 233–241. Available at: https://doi.org/10.59697/jik.v7i2.148.

Kusuma, I.H. and Cahyono, N. (2023) ‘Analisis Sentimen Masyarakat Terhadap Penggunaan E-Commerce Menggunakan Algoritma K-Nearest Neighbor’, 8(3), pp. 302–307.

Nandaresta, S.C. and Warman, C. (2023) ‘Terhadap Tiktok Shop Dan Shopee Di Twitter Menggunakan Metode Naïve Bayes Dan Knn ( K- Nearest Neighbor’, Sismatik, 12(1), pp. 1–9.

Permana, A.A., Noviyanto, W.A. and Kristiyanti, D.A. (2023) ‘Sentimen Analisis Opini Masyarakat Terhadap UMKM Pada Media Sosial Twitter Dengan Metode Naïve Bayes Classifier’, 12, pp. 163–170.

Tanniewa, A.M. (2023) ‘Implementasi Algoritma Support Vector Learning Terhadap Analisis Sentimen Penggunaan Aplikasi Tiktok Shop Seller Center’, pp. 165–170.




DOI: http://dx.doi.org/10.36499/jinrpl.v6i1.10313

Refbacks

  • There are currently no refbacks.


INDEXED BY :

Google Scholar GarudaCrossrefBASE Dimensions DOAJOne Search Scilit Sinta

 

 

 


Address : :

Fakultas Teknik Universitas Wahid Hasyim

JL. Menoreh Tengah X / 22, Sampangan, Gajahmungkur, Kota Semarang, Jawa Tengah 50232, Indonesia
Handphone: 0815-6529-309
Email: jinformatika@unwahas.ac.id
 
 
RJI JournalStories Main logo
View My Stats
ISSN : 2656-2855   E-ISSN : 2685-5518