What type of study is this?

This is a Quantitative Study study.

October 22, 2025Open Access

Sentiment Analysis of the 2025 Budget Cut Policy using the SVM Method

Key Points

The sentiment analysis shows that negative sentiment dominates responses to Indonesia's budget cuts.
Using the SVM method, the model achieved an accuracy of 60% with training on 799 tweets.
Negative sentiment accounted for 42.54% of tweets, highlighting public discontent with budget policies.
The combined performance metrics indicate room for improvement in sentiment classification accuracy.

Abstract

Budget cuts are a government policy to reduce the allocation of funds to various sectors or specific programs to align with state spending priorities. President Prabowo Subianto's efforts to streamline the budget remain controversial. The budget cuts are outlined in Presidential Instruction Number 1 of 2025. In the 2025 Fiscal Year, the budget cuts reached IDR 256.1 trillion. This study aims to analyze the Indonesian public's response to the 2025 budget efficiency policy via Twitter. The data used in this analysis are tweets discussing the policy. The data will then be classified into three sentiment categories: Negative, Neutral, and Positive. The results of Sentiment Analysis using the Support Vector Machine (SVM) method show that the model with 799 training data and 200 test data achieved an accuracy of 60.00%. The Negative class has a recall of 0.65, a precision of 0.66, and an f1-score of 0.65. The Neutral class has a precision of 0.45, a recall of 0.60, and an f1-score of 0.51, while the Positive class has a recall of 0.54 and an f1-score of 0.61. The macro averages of precision, recall, and f1-score are 0.60, 0.60, and 0.59, respectively, indicating that the performance has not yet reached its maximum. From the sentiment distribution graph, Negative sentiment dominates with 42.54% (425 tweets), followed by Positive 32.43% (324 tweets), and Neutral 25.03% (250 tweets).

Ask AI

Helpful

Bookmark

View Full Paper