ANALISIS DATA RESPONS PUBLIK PENGGUNA FITUR LIVE STREAMING TIKTOK SEBAGAI MEDIA HIBURAN DI INDONESIA DENGAN MENGGUNAKAN ALGORITMA NAIVE BAYES

Anisa Fadilah Saputri, . (2025) ANALISIS DATA RESPONS PUBLIK PENGGUNA FITUR LIVE STREAMING TIKTOK SEBAGAI MEDIA HIBURAN DI INDONESIA DENGAN MENGGUNAKAN ALGORITMA NAIVE BAYES. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (216kB)
[img] Text
AWAL.pdf
Download (638kB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (314kB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (403kB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (446kB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (6MB)
[img] Text
BAB 5.pdf
Download (233kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (228kB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (127kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository UPNVJ Only
Download (3MB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (23MB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (538kB)

Abstract

The TikTok live streaming feature is an interactive service that allows users to broadcast live and interact with audiences in real-time. This study aims to analyze user sentiment toward the TikTok live streaming feature to understand public perception, build a sentiment model using the Naive Bayes algorithm, and provide relevant recommendations for developers to improve the feature’s quality. Data were obtained from a questionnaire distributed to users of the TikTok live streaming feature in Indonesia, resulting in a total of 416 opinions categorized into two types of sentiment: positive and negative. The analysis process involved data preprocessing and labeling, resulting in 241 (57.9%) positive sentiments and 175 (42.1%) negative sentiments. Sentiment classification was performed using the Multinomial Naive Bayes algorithm and compared with K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) algorithms. Data were split into 60:40, 70:30, and 80:20 train-test ratios. The evaluation results showed that the Multinomial Naive Bayes algorithm consistently provided the best performance for classifying user sentiment toward the TikTok live streaming feature. The highest performance was achieved at the 80:20 train-test ratio, with 90% accuracy, 91% precision, 89% recall, and 90% f1-score. Meanwhile, the KNN and SVM algorithms showed stable performance in the 86%–88% accuracy range without significant improvement. This study demonstrates that the Naive Bayes algorithm is the most optimal choice for sentiment classification and effective in providing strategic recommendations to enhance the user experience of TikTok’s live streaming feature.

Item Type: Thesis (Skripsi)
Additional Information: [No. Panggil: 2110512010] [Pembimbing 1: Tjahjanto] [Pembimbing 2: Sarika] [Penguji 1: Bambang Saras Yulistiawan] [Penguji 2: Novi Trisman Hadi]
Uncontrolled Keywords: Sentiment Analysis, Live Streaming, TikTok, Naive Bayes, Classification
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Fakultas Ilmu Komputer > Program Studi Sistem Informasi (S1)
Depositing User: ANISA FADILAH SAPUTRI
Date Deposited: 30 Jul 2025 10:37
Last Modified: 15 Aug 2025 00:48
URI: http://repository.upnvj.ac.id/id/eprint/37510

Actions (login required)

View Item View Item