Nayandra Agastia Putra, . (2026) PERANCANGAN APLIKASI UNTUK PREDIKSI JUMLAH PENAYANGAN PADA AKUN KONTEN KREATOR TIKTOK. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
Abstrak.pdf Download (154kB) |
|
Text
AWAL.pdf Download (668kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (345kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (346kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (286kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
BAB 5.pdf Download (239kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (214kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (147kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (5MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (13MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (653kB) |
Abstract
The issue of uncertainty in content performance on TikTok social media has become a significant challenge for content creators in formulating effective strategies. Fluctuating video performance that is difficult to predict manually often hinders audience reach optimization. To address this, a web-based view count prediction application was designed and developed in this study with enhanced text analysis capabilities. The system utilizes the Random Forest Classifier algorithm combined with a Natural Language Processing (NLP) approach to process video statistical data and text features from captions. Data covering interaction metrics as well as content keywords undergoes preprocessing using the Sastrawi library, feature engineering, and labelling into Trending or Non-Trending categories before being used to train the model. The trained model is integrated into a system based on the Streamlit framework, which is now equipped with Batch Prediction features and an Analytics Dashboard. Results indicated that the integration of the 'Likes' metric and text features had a dominant influence, with the model achieving 93.37% accuracy. This system is intended to function as a practical and adaptive decision support tool, empowering creators to evaluate content virality potential efficiently prior to upload.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2110512051] [Pembimbing 1: Iin Ernawati] [Pembimbing 2: Nindy Irzavika] [Penguji 1: Andhika Octa Indarso] [Penguji 2: Catur Nugrahaeni Puspita Dewi] |
| Uncontrolled Keywords: | TikTok, Views Prediction, Random Forest Algorithm, Natural Language Processing (NLP), Streamlit, Batch Prediction |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Sistem Informasi (S1) |
| Depositing User: | NAYANDRA AGASTIA PUTRA |
| Date Deposited: | 26 Mar 2026 08:26 |
| Last Modified: | 26 Mar 2026 08:26 |
| URI: | http://repository.upnvj.ac.id/id/eprint/42174 |
Actions (login required)
 |
View Item |