Alfansuriy Rahmanda Ikhsan, Norma Juainah, Najmi Muhammad Fadli

Analisis Sentimen Pada Postingan Akun @Hermanderu67 Dan Influencer Palembang Di Instagram Selama Pilgub 2024 Dalam Menggunakan Klasifikasi Naive Bayes

Introduction

Analisis sentimen pada postingan akun @hermanderu67 dan influencer palembang di instagram selama pilgub 2024 dalam menggunakan klasifikasi naive bayes. Analisis sentimen postingan Instagram @hermanderu67 dan influencer Palembang selama Pilgub Sumsel 2024 menggunakan Naive Bayes. Temukan sentimen publik dan pengaruh media sosial pada opini pemilih.

Abstract

This research aims to analyze public sentiment on Instagram posts from the @hermanderu67 account and several Palembang influencers during the 2024 South Sumatra gubernatorial election. Using the Naïve Bayes classification method, this study examines user comments to identify the tendencies of positive, negative, and neutral sentiments toward the political messages conveyed. The analysis is based on the growing phenomenon of social media as a political campaign arena, particularly among young voters, where public opinion is largely shaped through rapid and massive digital interactions.The findings indicate that Herman Deru’s political communication strategy through social media was relatively successful in building a positive image, as reflected by the dominance of positive sentiments in user comments. However, the presence of negative and neutral sentiments also reveals that social media is a dynamic space of political discourse, where criticism and appreciation coexist. This study affirms that social media functions not only as a channel for information distribution but also as a digital political arena that significantly influences public opinion and voter preferences.

Review

This study delves into the critical area of political sentiment analysis on Instagram, focusing specifically on the 2024 South Sumatra gubernatorial election. By examining posts from the @hermanderu67 account and selected Palembang influencers, the research aims to classify public sentiment (positive, negative, neutral) using the Naïve Bayes method. The relevance of this work is particularly high, given the increasing role of social media platforms, especially Instagram, in shaping political discourse and influencing voter behavior, particularly among younger demographics. This timely investigation offers insights into how digital interactions contribute to public opinion formation during electoral campaigns. A significant strength of this research lies in its empirical application to a real-world political event, providing valuable data on the dynamics of online sentiment within a specific regional election context. The use of Naïve Bayes, a a well-established classification algorithm, is appropriate for this type of text-based sentiment analysis, offering a quantifiable approach to understanding public perception. The findings are particularly insightful, indicating the relative success of Herman Deru's social media strategy in cultivating a positive image, while simultaneously acknowledging the complex and dynamic nature of social media where both positive and critical feedback coexist. This work effectively reinforces the notion of social media as a powerful digital political arena, extending beyond mere information dissemination to actively influencing voter preferences. While the abstract presents a compelling overview and promising results, a full paper would benefit from greater detail on several methodological aspects. Clarification on the selection criteria for "Palembang influencers" and the methodology for data collection (e.g., comment sampling, period, total volume) would enhance the study's replicability and generalizability. Furthermore, discussing potential limitations of the Naïve Bayes classifier in handling nuances of political language, sarcasm, or evolving slang, along with strategies employed to mitigate these, would strengthen the analysis. Exploring the demographic profile of the commenters, if feasible, could add another layer of depth to understanding voter sentiment. Overall, this research provides a solid foundation for understanding digital political communication, and further exploration of these details would significantly enrich its contribution to the field.

Full Text

Comments

You need to be logged in to post a comment.