Skripsi

Sentiment analysis of corruption eradication efforts in 2024 during president prabowo’s leadership using machine learning / Heni Susilawati

Abstrak
This study examines public sentiment towards corruption eradication efforts under the leadership of President Prabowo in 2024 by utilizing machine learning techniques. The data used was obtained from the YouTube platform and grouped into three sentiment categories positive negative and neutral. The labeling process is done in two ways namely manually and automatically using a lexicon dictionary. Two machine learning algorithms namely Support Vector Machine (SVM) and Random Forest were applied to classify the sentiments. The results show that the automatically labeled sentiment analysis achieves high accuracy for both algorithms at 98%. However the models had difficulty in recognizing positive and negative sentiments while their performance in detecting neutral sentiments was satisfactory. In contrast manual labeling resulted in lower accuracy 82% for SVM and 79% for Random Forest. Nonetheless this accuracy is still quite good for these algorithms. These findings highlight the importance of choosing a labeling method for sentiment analysis especially on complex topics such as corruption eradication.

Informasi Detail

DDC

SKRIPSI DIGITAL

Prodi

Universitas Negeri Malang. Program Studi Matematika, 2025.

Deskripsi Fisik

xii, 77 hlm. : ilus.

Bahasa

No Reg

2965/RS/25

Edisi

Skripsi (Sarjana)--Universitas Negeri Malang. 2025

Subjek

1. MACHINE LEARNING - PEMBERANTASAN KORUPSI
2. MACHINE LEARNING - CORRUPTION ERADICATION

Pembimbing

1. Dahliatul Hasanah, S.si, M.math.sc.

Lampiran Berkas