PERBANDINGAN METODE SUPPORT VECTOR MACHINE DAN NAÏVE BAYES DENGAN LAPLACE ESTIMATOR DALAM KASUS ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP KURIKULUM MERDEKA BELAJAR KAMPUS MERDEKA (MBKM)

AMELIA ARIEFAH, . (2023) PERBANDINGAN METODE SUPPORT VECTOR MACHINE DAN NAÏVE BAYES DENGAN LAPLACE ESTIMATOR DALAM KASUS ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP KURIKULUM MERDEKA BELAJAR KAMPUS MERDEKA (MBKM). Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (777kB)
	Text BAB 1.pdf Download (506kB)
	Text BAB 2.pdf Restricted to Registered users only Download (846kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (665kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (759kB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (486kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (416kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (844kB) | Request a copy

Abstract

Pergantian kurikulum sangatlah berdampak pada bidang pendidikan di Indonesia, khususnya di perguruan tinggi. Hal ini dikarenakan kurikulum digunakan sebagai sarana dalam upaya mencapai tujuan keberhasilan pendidikan sekaligus pedoman dalam pelaksanaan pengajaran pada setiap tingkat pendidikan. Penelitian ini bertujuan untuk mengetahui sentimen publik mengenai penerapan dari kurikulum Merdeka Belajar Kampus Merdeka (MBKM) melalui media sosial Twitter serta mengetahui perbandingan kinerja dari metode Support Vector Machine dan Naïve Bayes dengan Laplace Estimator. Analisis sentimen dilakukan dengan melihat kecenderungan opini terhadap suatu masalah atau objek lebih ke arah negatif atau positif. Penelitian ini menggunakan metode Support Vector Machine dan Naïve Bayes dengan Laplace Estimator terhadap hasil pencarian tweets dengan kata kunci “kurikulum mbkm” dan “mbkm”. Didapatkan sebanyak 2500 data tweets yang akan dibagi menjadi 2000 data sebagai data training dengan rincian 1000 tweet positif dan 1000 tweet negatif serta 500 data sebagai data testing. Hasil penelitian menunjukkan bahwa metode Support Vector Machine mencapai nilai tertinggi di semua metrik yaitu nilai accuracy sebesar 83.00%, nilai precision sebesar 85.00%, nilai recall 80.00%, dan nilai f1 score 83.00%. ----- Curriculum changes have a significant impact on the field of education in Indonesia, especially in universities. This is because the curriculum is used as a means in an effort to achieve the goals of educational success as well as guidelines in the implementation of teaching at every level of education. This study aims to determine public sentiment regarding the application of the Merdeka Belajar Kampus Merdeka (MBKM) curriculum through Twitter social media and find out the comparison of performance from the Support Vector Machine method and the combination of Naïve Bayes with Laplace Estimator. This study used the Support Vector Machine method and the combination of Naïve Bayes with Laplace Estimator on the search results of tweets with the keywords "kurikulum mbkm" and "mbkm". The data used in this study amounted to 2500 data tweets which will be divided into 2000 data as training data with details of 1000 positive tweets and 1000 negative tweets and 500 data as testing data. The results showed that the performance of the Support Vector Machine method reached the highest value in all metrics, resulted in an accuracy value of 83.00%, precision value of 85.00%, recall value of 80.00%, and f1 score value of 83.00% when compared to the combined performance of the Naïve Bayes with Laplace Estimator which resulted in an accuracy value of 80.20%, precision value of 82.00%, recall value of 79.00%, and f1 score value of 80.00%.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Widodo, S.Kom., M.Kom. ; 2). Murien Nugraheni, S.T., M.Cs.
Subjects:	Teknologi dan Ilmu Terapan > Teknik Komputer
Divisions:	FT > S1 Pendidikan Teknik Informatika Komputer
Depositing User:	Users 18144 not found.
Date Deposited:	31 Jul 2023 04:42
Last Modified:	16 Jan 2024 00:54
URI:	http://repository.unj.ac.id/id/eprint/38914

Actions (login required)

View Item