ANALISIS SENTIMEN KEBIJAKAN PENGHAPUSAN KEWAJIBAN SKRIPSI PERGURUAN TINGGI PENGGUNA INSTAGRAM MENGGUNAKAN METODE LOGISTIC REGRESSION

FATHIA DWI ASTUTI, . (2024) ANALISIS SENTIMEN KEBIJAKAN PENGHAPUSAN KEWAJIBAN SKRIPSI PERGURUAN TINGGI PENGGUNA INSTAGRAM MENGGUNAKAN METODE LOGISTIC REGRESSION. Sarjana thesis, UNIVERSITAS NEGERI JAKART.

	Text COVER.pdf Download (4MB)
	Text BAB 1.pdf Download (1MB)
	Text BAB 2.pdf Restricted to Registered users only Download (427kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (465kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (454kB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (806kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (1MB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Sistem pendidikan Indonesia menjadikan skripsi sebagai salah satu syarat kelulusan perguruan tinggi untuk memperoleh gelar. Namun pada tanggal 29 Agustus 2023, Menteri Pendidikan, Kebudayaan, Riset, dan Teknologi menyampaikan kebijakan baru mengenai skripsi yang tidak lagi diwajibkan dan dapat diganti dalam bentuk lain, kebijakan ini diusut dalam Peraturan Menteri Pendidikan, Kebudayaan, Riset, dan Teknologi Nomor 53 Tahun 2023 tentang Penjaminan Mutu Pendidikan Tinggi. Penelitian ini bertujuan untuk melakukan analisis sentimen kebijakan peghapusan kewajiban skripsi perguruan tinggi pada komentar Instagram dan mengevaluasi kinerja klasifikasinya. Analisis sentimen dilakukan dengan dua skenario menggunakan metode Logistic Regression untuk klasifikasi, pembobotan Term Frequency-Inverse Document Frequency (TF-IDF), validasi data 10-fold Cross Validation, oversampling data Adaptive Synthetic (ADASYN) dan evaluasi kinerja Confusion Matrix. Skenario pertama menghasilkan accuracy 87,5%, precision 88,2%, recall 90,5%, dan f-1 score 89,4% dan skenario kedua menghasilkan accuracy 89,6%, precision 91,5%, recall 87,7%, dan f-1 score 89,5%. Dari evaluasi kedua skenario dapat disimpulkan bahwa model klasifikasi dengan metode Logistic Regression memiliki kinerja yang baik pada tugas analisis sentimen kebijakan penghapusan kewajiban skripsi, terlebih lagi jika dikombinasikan dengan ADASYN. ; ***** The Indonesian education system requires a thesis as one of the requirements for university graduation to obtain a degree. However, on August 29 2023, the Minister of Education, Culture, Research and Technology announced a new policy regarding theses which are no longer required and can be replaced in another form, this policy is regulated in Minister of Education, Culture, Research and Technology Regulation Number 53 of 2023 concerning Quality Assurance of Higher Education. This study aims to analyze the sentiment of the policy of eliminating university thesis obligations on Instagram comments and evaluate its classification performance. Sentiment analysis was carried out with two scenarios using the Logistic Regression method for classification, Term Frequency-Inverse Document Frequency (TF-IDF) for weighting, 10-fold Cross Validation for data validation, Adaptive Synthetic (ADASYN) for oversampling data and Confusion Matrix for performance evaluation. The first scenario produces an accuracy of 87.5%, precision 88.2%, recall 90.5%, and f-1 score 89.4% and the second scenario produces an accuracy of 89.6%, precision 91.5%, recall 87.7 %, and f-1 score 89.5%. From the evaluation of the two scenarios, it can be concluded that the classification model using the Logistic Regression method has good performance in the task of sentiment analysis for the policy of eliminating thesis obligations, especially when combined with ADASYN.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Widodo, S.Kom, M.Kom. ; 2). Bambang Prasetya Adhi, S.Pd., M.Kom.
Subjects:	Teknologi dan Ilmu Terapan > Teknik Komputer
Divisions:	FT > S1 Pendidikan Teknik Informatika Komputer
Depositing User:	Users 23218 not found.
Date Deposited:	25 Jul 2024 05:44
Last Modified:	25 Jul 2024 05:44
URI:	http://repository.unj.ac.id/id/eprint/46548

Actions (login required)

View Item