Analisis Sentimen Tiktok Shop Pada Twitter Menggunakan Metode Multinomial Naïve Bayes Dengan Pembobotan Fitur BM25

Yasin, M. Andrew Arjunanda (2024) Analisis Sentimen Tiktok Shop Pada Twitter Menggunakan Metode Multinomial Naïve Bayes Dengan Pembobotan Fitur BM25. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (COVER) 20083010014_COVER.pdf Download (1MB)
	Text (BAB I) 20083010014_BAB I.pdf Download (50kB)
	Text (BAB II) 20083010014_BAB II.pdf Restricted to Repository staff only Download (264kB) \| Request a copy
	Text (BAB III) 20083010014_BAB III.pdf Restricted to Repository staff only until 30 May 2026. Download (4MB) \| Request a copy
	Text (BAB IV) 20083010014_BAB IV.pdf Restricted to Repository staff only until 30 May 2026. Download (1MB) \| Request a copy
	Text (BAB V) 20083010014_BAB V.pdf Download (36kB)
	Text (DAFTAR PUSTAKA) 20083010014_DAFTAR PUSTAKA.pdf Download (181kB)
	Text (LAMPIRAN) 20083010014_LAMPIRAN.pdf Restricted to Repository staff only Download (520kB) \| Request a copy

Abstract

Advancements in internet technology have significantly altered how users in Indonesia interact, particularly in commerce and social communication. One platform leveraging this advancement is TikTok with its TikTok Shop feature, which allows users to shop without leaving the app. However, TikTok Shop was temporarily closed on October 4, 2023, to comply with online trading regulations, before reopening on December 12, 2023. This situation sparked various responses on Twitter due to concerns about potential trade monopolies, necessitating sentiment analysis to understand public opinion. One sentiment analysis method is Multinomial Naïve Bayes, which calculates probabilities. This research process includes data collection from Twitter using the Python library "tweet harvest", totaling 1413 data points, data preprocessing, data labeling, term weighting with BM25 and TF-IDF, feature selection, model validation, classification using Multinomial, Gaussian, and Bernoulli Naïve Bayes methods, and word cloud visualization. The research aims to help the government make more informed policy decisions and provide insights for the public to respond wisely to the situation. The results indicate that Multinomial Naïve Bayes with BM25 achieves the highest accuracy of 0.75, with the majority of responses showing negative sentiment.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Prasetya, Dwi Arman	NIDN0005128001	arman.prasetya.sada@upnjatim.ac.id
Thesis advisor	Fahrudin, Tresna Maulana	NIDN0701059301	tresna.maulana.ds@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

M. Andrew Arjunanda Yasin

Date Deposited:

31 May 2024 08:54

Last Modified:

31 May 2024 08:54

URI:

https://repository.upnjatim.ac.id/id/eprint/23622