CD Skripsi

Analisis sentimen masyarakat pada media sosial twitter (x) terhadap isu keikutsertaan indonesia dalam ktt brics menggunakan Convolutional Neural Network (CNN)

---

This study aims to analyze Indonesian public sentiment regarding the issue of Indonesia's participation in the BRICS Summit through the social media platform Twitter (X). The method used in this research is the Convolutional Neural Network (CNN) for sentiment classification, combined with Fasttext as the word embedding technique. Data were collected through web scraping, resulting in 2,025 tweets, which after cleaning and preprocessing yielded 1,158 usable entries. These were categorized into three sentiment classes: positive, negative, and neutral. To address class imbalance, an oversampling technique was applied. The evaluation results show that without oversampling, the model achieved only 54% accuracy, while after oversampling, the accuracy increased to 81%. The model performance was evaluated using accuracy, precision, recall, and F1-score. This study demonstrates that the combination of Fasttext and CNN is effective for analyzing public opinion on strategic issues such as Indonesia’s participation in BRICS.

Keywords: Sentiment Analysis, BRICS, CNN, Fasttext, Twitter, Oversampling

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103125620

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103125620

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar