CD Skripsi

Analisis Sentimen Ulasan Mahasiswa Pada Sistem Edom Menggunakan Model Bahasa Berbasis Transformer

---

Reviews and opinions from users play a crucial role in enhancing the quality and performance of various services and platforms, including higher education institutions. Student reviews can be used as an evaluation material to improve the quality of teaching, campus facilities, and the overall learning experience. The EDOM system is an online platform that facilitates students to provide reviews about their learning experiences. However, this system is not yet equipped with an automatic review analysis system. Manual analysis of student reviews can consume significant time and effort, as it requires human labor to read and classify each review. This study proposes the use of Transformer-based language models for sentiment analysis. Transformer-based language models are a type of neural network capable of learning the relationship between words in a sentence regardless of word order. This study tested four different Transformer-based language models, namely M-BERT, IndoBERT, RoBERTa Indonesia, and GPT-2 Indonesia. The dataset used in this study consisted of 31,600 student reviews from the EDOM system at Riau University, which were then labeled as positive or negative. The research results show that the IndoBERT-base-uncased model performed the best, with an MCC value of 88.2%, accuracy of 94.1%, precision of 94.5%, recall of 93.2%, and f1-score of 93.8%. Based on these results, it can be concluded that the IndoBERT model is the most effective Transformer-based language model for sentiment analysis of student reviews.

Keywords: Machine Learning, Natural Language Processing, Sentiment Analysis, Transformer, EDOM

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003113132

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003113132

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