Claim Missing Document
Check
Articles

Found 22 Documents
Search

Tinjauan Analisis Sentimen Terkait COVID-19 Al Sidqi, Muhammad Affan; Afrizal, Nabil Nur; Rodiansyah, Novan; Cahyana, Rinda
Journal of Digital Literacy and Volunteering Vol. 3 No. 1 (2025): January
Publisher : Puslitbang Akademi Relawan TIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57119/litdig.v3i1.118

Abstract

The Covid-19 pandemic control policy has created public opinion on social media. The government controls sentiment so that people remain compliant with the policy. Several previous studies have analyzed sentiment with various approaches. This study aims to describe how to analyze sentiment related to the pandemic of earlier studies with a traditional literature review approach through the literature survey stage. The results of the review found various approaches to data collection and sentiment analysis that have been applied by previous studies and their challenges, as well as opportunities for further research.
Analisis Sentimen Ulasan Wisata Budaya Menggunakan Metode Support Vector Machine dan Long Short-Term Memory Ramdani, Rizki; Cahyana, Rinda
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2585

Abstract

In the era of digital transformation, tourist behavior in expressing perceptions of travel destinations has increasingly shifted toward online platforms such as Google Maps and Twitter. These digital reviews not only represent individual experiences but also reflect collective opinions that can serve as a foundation for formulating data-driven tourism development policies. This study aims to conduct sentiment analysis on public opinion regarding Kampung Naga by comparing the performance of two classification algorithms: Support Vector Machine (SVM) and Long Short-Term Memory (LSTM). The methodological approach employed is SEMMA (Sample, Explore, Modify, Model, Assess). The dataset comprises 2,469 reviews obtained through web scraping techniques from Google Maps and Twitter. All data underwent preprocessing stages including cleaning, tokenization, stopword removal, and automatic sentiment labeling using the ChatGPT language model, with three classification labels: positive, neutral, and negative. Modeling was performed using SVM with TF-IDF representation and LSTM with an embedding layer. Model evaluation utilized precision, recall, and F1-score metrics. The results indicate that SVM achieved an accuracy of 83% and performed best on neutral sentiment, while LSTM recorded an accuracy of 81% with stable performance on positive and neutral sentiments. This research contributes to the development of text-based public opinion analysis systems to support the promotion and management of cultural tourism destinations.