Sentiment Analysis of the Kampus Merdeka Program on Twitter - Using Support Vector Machine and Feature Extraction Comparison, TF-IDF vs FastText

Sep 10, 2026

Kampus Merdeka, launched by Indonesia’s Ministry of Education, Culture, Research, and Technology in 2020, seeks to strengthen students’ skills through practical work experience. Given the growing importance of social media for measuring public opinion, this study analyzes sentiment toward Kampus Merdeka on Twitter. Tweets are classified into positive, negative, or neutral sentiments using Support Vector Machine, while comparing TF-IDF and FastText for text-to-vector representation. Results from 16,730 tweets show 73% accuracy with FastText versus 72% with TF-IDF, with predominantly positive sentiment (60.7%).

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Introduction

Program background and public attention on Twitter

Problem statement and need for measurable sentiment understanding

Research gap and study objective

Proposed approach and modeling plan

Sentiment Analysis of the Kampus Merdeka Program on Twitter Using Support Vector Machine and a Feature Extraction Comparison: TF-IDF vs. FastText Lasmedi Afuan 1,*, , Nurul Hidayat2,, Nofiyati3, Mohamad Faris As'ad4 1,2,3,4Department of Informatics, Engineering Faculty, Universitas Jenderal Soedirman, Indonesia (Received: July 19, 2024; Revised: August 13, 2024; Accepted: September 11, 2024; Available online: October 15, 2024) Abstract The Kampus Merdeka program, launched by the Indonesian Ministry of Education, Culture, Research, and Technology in 2020, aims to enhance students' skills through hands-on work experience. Considering the rising significance of social media, particularly Twitter, in gauging public opinion, this research focuses on analyzing the sentiment towards the Kampus Merdeka program. The primary objective is to classify the sentiments expressed in tweets related to the program and compare two feature extraction techniques—TF-IDF and FastText—to identify the best approach for transforming text data into numerical vectors. The sentiment classification model was built using the Support Vector Machine (SVM) algorithm, a machine learning technique known for its accuracy in text classification. A total of 16,730 tweets were collected and analyzed, yielding an accuracy of 73% for FastText and 72% for TF-IDF. Results show that FastText is more effective in capturing semantic relationships, leading to higher accuracy in sentiment classification. Findings indicate that the public sentiment towards the Kampus Merdeka program is predominantly positive (60.7%), with negative and neutral sentiments at 33.5% and 5.8%, respectively. The success of the FastText method underscores the importance of advanced feature extraction techniques in text classification. The novelty of this research lies in its use of FastText for educational policy evaluation, providing a new perspective on using sentiment analysis to assess public perception of educational programs. Keywords: Sentiment Analysis, FastText, Kampus Merdeka, Support Vector Machine, TF-IDF, Twitter 1. Introduction The transformation of higher education in Indonesia has become a strategic agenda to develop superior human resources that are globally competitive [1], [2] . One of the key initiatives launched by the Ministry of Education, Culture, Research, and Technology (Kemendikbudristek) in 2020 is the Kampus Merdeka program [3] . This program offers students opportunities to gain practical work experience through internships, independent studies, and teaching. Since its launch, the program has garnered significant public attention, especially among students and educators, who have voiced their opinions on social media platforms such as Twitter. These varied reactions present an opportunity to analyze public sentiment to better understand the perceptions surrounding the Kampus Merdeka program. However, the issue arises from the lack of structured, measurable understanding of how the public, especially on Twitter, perceives the Kampus Merdeka program. Twitter, being one of the most popular social media platforms in Indonesia, provides a public space for open expression [4]. Nevertheless, transforming thousands, or even millions, of tweets into interpretable data requires appropriate analytical approaches. This is where sentiment analysis becomes crucial, enabling the categorization of public opinions into positive, negative, or neutral sentiments. This research is conducted because of the scarcity of prior studies specifically addressing public sentiment toward the Kampus Merdeka program using text mining and machine learning approaches. Although previous studies have applied sentiment analysis to various other issues, there is a gap in comprehensive research that leverages this technology to measure public perceptions of educational policies in Indonesia, particularly this program. This knowledge gap must be filled to provide insights for policymaker

FAQs

What does the research analyze regarding Kampus Merdeka?

It classifies public sentiment toward the Kampus Merdeka program based on tweets from Twitter into positive, negative, and neutral categories.

Which machine learning method is used for sentiment classification?

The model uses Support Vector Machine (SVM) for text sentiment classification.

How do TF-IDF and FastText compare in performance?

FastText achieves 73% accuracy, slightly higher than TF-IDF at 72%, indicating better semantic relationship capture.

Sep 10, 2026

Uploaded by Oliver Hayes