Adrianto, Muhammad Tengku (2025) Analisis Sentimen Calon Gubernur Riau Pada Pilkada 2024 Berdasarkan Tweet Pada Media Sosial X Menggunakan Metode Naive Bayes dan Support Vector Machine (SVM). Other thesis, Universitas Islam Riau.

Text skripsi_203510541_watermark.pdf - Submitted Version Restricted to Registered users only Download (6MB) | Request a copy

Abstract

Regional Head Election (Pilkada) is one of the crucial democratic processes in determining the direction of regional government, especially in Riau. Pilkada is not only an event for candidates to express their vision and mission, but also functions as a space for the community to convey their assessments and opinions on prospective leaders. Social media, especially applications such as Twitter (X), is a platform where people often share their feelings and views regarding the 2024 Riau Pilkada. This study aims to analyze the sentiment of X users towards the 2024 Riau Pilkada among the Riau community using tweets uploaded on the X platform. In this study, testing was carried out by comparing Machine Learning algorithms, namely Naive Bayes and Support Vector Machine (SVM) to classify positive and negative sentiments. The dataset used was 5000 with the number of negative sentiments 2800 and positive sentiments 2200. The data was trained and tested using a ratio of 90:10, 80:20, 70:30 on both algorithms with the Bag of Words (BoW) approach. The highest accuracy result is in Support Vector Machine (SVM) with a ratio of 80:20, which is 83.5%. In Naive Bayes itself, it gets 69.8% with a ratio of 90:10.

Actions (login required)

View Item