Indonesian Journal of Applied Informatics
Vol 9, No 1 (2024)

Analisis Sentimen Pasar melalui Berita Finansial untuk Prediksi Harga Saham PT Bank Rakyat Indonesia Tbk

Ferdyansyah Permana Putra (Insitut Teknologi Sepuluh Nopember)
Mukti Ratna Dewi (Insitut Teknologi Sepuluh Nopember)
Fausania Hibatullah (Institut Teknologi Sepuluh Nopember)



Article Info

Publish Date
23 Nov 2024

Abstract

Abstrak:Sentimen pasar merupakan salah satu faktor yang mempengaruhi fluktuasi harga saham yang dapat bersumber dari masyarakat umum maupun berita-berita yang terkait dengan saham. Dalam penelitian ini, pengaruh sentimen pasar melalui berita keuangan dianalisis terhadap harga saham PT Bank Rakyat Indonesia Tbk (BBRI). Penelitian ini menggunakan pendekatan machine learning dengan metode Support Vector Regression (SVR) untuk memprediksi harga penutupan saham BBRI berdasarkan sentimen berita. Model SVR dioptimalkan dengan algoritma Fruit Fly Optimization Algorithm (FOA). Sentimen pasar terlebih dahulu dievaluasi menggunakan metode IndoBERT yang menunjukkan tingkat akurasi sentimen keseluruhan di atas 90%. Setelah itu, empat skenario pemodelan diusulkan untuk menemukan model prediksi terbaik: (1) model tanpa sentimen, (2) model dengan sentimen pada periode , (3) model dengan sentimen pada periode , dan (4) model dengan sentimen pada periode  dan periode . Hasil akhir menunjukkan bahwa model pada skenario (1) memiliki kesalahan prediksi terendah dibandingkan dengan model lainnya==============================================Abstract:Market sentiment is one of the factors that influences the fluctuation of stock prices, which can originate from the general public or news related to stocks. In this study, we explore the effect of market sentiment through financial news on the stock price of PT Bank Rakyat Indonesia Tbk (BBRI). This research adopts a machine learning approach using the Support Vector Regression (SVR) method to predict the closing price of BBRI stock based on news sentiment, and the function is later optimized with the Fruit Fly Optimization Algorithm (FOA) algorithm. The market sentiment is first evaluated using the IndoBERT method, which shows an overall sentiment accuracy level above 90%. Afterward, four modeling scenarios are proposed to find the best prediction model: (1) a model without sentiment, (2) a model with sentiment at period , (3) a model with sentiment at period , and (4) a model with sentiment at both period  and period . The final results indicate that the model in scenario (1) has the lowest prediction error compared to other models

Copyrights © 2024






Journal Info

Abbrev

ijai

Publisher

Subject

Computer Science & IT

Description

Indonesian Journal of Applied Informatics publishes articles that are of significance in their respective fields whilst also contributing to the discipline of informatics as a whole and its application. Every incoming manuscript will first be examined by the Editorial Board in accordance with ...