Skripsi

ANALISIS SENTIMEN TERHADAP CRYPTOCURRENCY MENGGUNAKAN MODEL BERT

---

The rapid development of digital technology has driven various innovations in the financial sector, one of which is cryptocurrency, which is increasingly discussed on social media and various digital platforms. On the other hand, price volatility, security issues, and evolving regulations have led to diverse public opinions regarding cryptocurrency. Therefore, this study aims to analyze public sentiment toward cryptocurrency using the BERT model, focusing on positive and negative sentiments. The dataset used in this research was obtained from the Crypto News and Bitcoin Tweets datasets available on Kaggle, consisting of 54,208 data entries, including 35,904 positive sentiments and 18,304 negative sentiments. The data were first processed through a preprocessing stage and then continued with the finetuning process of the BERT model using eight experimental scenarios. The results show that the best-performing model was obtained in Scenario 9 with a learning rate of 2e-5, max length 60 and 3 epochs, achieving an accuracy of 90.08%, precision of 90.07%, recall of 90.08%, and an F1-score of 90.07%.

---

Availability

Detail Information

Series Title

-

Call Number

T1960122026

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xvi, 164 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Analisis sentimen
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment