ANALISIS SENTIMEN PUBLIK TERKAIT CPNS DAN PPPK MENGGUNAKAN ALGORITMA MACHINE LEARNING PADA KOMENTAR MEDIA SOSIAL KEMENTERIAN PANRB

Abas, Sutirto (2026) ANALISIS SENTIMEN PUBLIK TERKAIT CPNS DAN PPPK MENGGUNAKAN ALGORITMA MACHINE LEARNING PADA KOMENTAR MEDIA SOSIAL KEMENTERIAN PANRB. Sarjana thesis, Universitas Muhammadiyah Gorontalo.

[img] Text (Abstrak Sutirto Abas)
FILE ABSTRAK.pdf

Download (536kB)
[img] Text (Bab I Sutirto Abas)
FILE BAB 1.pdf

Download (251kB)
[img] Text (Daftar Pustaka Sutirto Abas)
FILE DAFTAR PUSTAKA.pdf

Download (246kB)
[img] Text (Skripsi Sutirto Abas)
SKRIPSI - SUTIRTO ABAS.pdf
Restricted to Repository staff only

Download (2MB)

Abstract

The postponement of the appointment of Civil Service Candidates (CPNS) and Government Employees with Work Contracts (PPPK) has sparked a variety of public reactions, many of which have been expressed on the social media platform X. This study aims to analyze patterns of public sentiment toward this policy and to compare the performance of the Naive Bayes and Random Forest algorithms in classifying public sentiment. The study employs a quantitative approach using an experimental method based on the Cross Industry Standard Process for Data Mining (CRISP-DM) framework. Data were collected via web scraping on the social media platform X using keywords related to CPNS and PPPK during the period of March 5–16, 2025, with an initial total of 1,423 comments, which was reduced to 1,180 data points after duplicate removal. The preprocessing stages included cleaning, case folding, normalization, tokenization, and stopword removal. Sentiment labeling was performed using the InSet dictionary, while feature extraction used the TF-IDF method. The model was built using the Naive Bayes and Random Forest algorithms with a data split of 80% for training and 20% for testing. Model performance evaluation was performed using a confusion matrix and ROC curve. The results of the study indicate that public sentiment is dominated by negative sentiment, accounting for 61.27%. The Random Forest model achieved an accuracy of 78% with an AUC of 0.88, outperforming the Naive Bayes model, which had an accuracy of 75% and an AUC of 0.86. These results suggest that the Random Forest model is more effective at classifying sentiment in Indonesian language social media text data.

Item Type: Thesis (Sarjana)
Contributors:
ContributionContributorsNIDN/NBMEmail
Thesis advisorAbas, Mohammad Ilyas0926089101[email protected]
Thesis advisorPranata, Widya Eka1626119801[email protected]
Additional Information: Penundaan pengangkatan Calon Pegawai Negeri Sipil (CPNS) dan Pegawai Pemerintah dengan Perjanjian Kerja (PPPK) menimbulkan beragam respons publik yang banyak disampaikan melalui media sosial X. Penelitian ini bertujuan untuk menganalisis pola sentimen publik terhadap kebijakan tersebut serta membandingkan kinerja algoritma Naive Bayes dan Random Forest dalam mengklasifikasikan sentimen masyarakat. Penelitian menggunakan pendekatan kuantitatif dengan metode eksperimen berbasis kerangka kerja Cross Industry Standard Process for Data Mining (CRISP-DM). Data dikumpulkan melalui teknik web scraping pada media sosial X dengan kata kunci terkait CPNS dan PPPK selama periode 5–16 Maret 2025 dengan jumlah awal 1.423 komentar, yang setelah proses penghapusan duplikasi menjadi 1.180 data. Tahapan preprocessing meliputi cleaning, case folding, normalisasi, tokenisasi, dan stopword removal. Pelabelan sentimen dilakukan menggunakan kamus InSet, sedangkan ekstraksi fitur menggunakan metode TF-IDF. Model dibangun menggunakan algoritma Naive Bayes dan Random Forest dengan pembagian data 80% training dan 20% testing. Evaluasi kinerja model dilakukan menggunakan confusion matrix dan kurva ROC. Hasil penelitian menunjukkan bahwa sentimen publik didominasi oleh sentimen negatif sebesar 61,27%. Random Forest memperoleh akurasi sebesar 78% dengan nilai AUC 0,88, lebih tinggi dibandingkan Naive Bayes dengan akurasi 75% dan AUC 0,86. Hasil ini menunjukkan bahwa Random Forest lebih optimal dalam mengklasifikasikan sentimen pada data teks media sosial berbahasa Indonesia.
Uncontrolled Keywords: Analisis Sentime, CPNS, PPPK, Naive Bayes, Random Forest
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Divisions: Fakultas Sains dan Ilmu Komputer > Program Studi Ilmu Komputer
Depositing User: Sutirto Abas
Date Deposited: 09 Jul 2026 06:58
Last Modified: 09 Jul 2026 06:58
URI: http://repository.umgo.ac.id/id/eprint/6729

Actions (login required)

View Item View Item