ANALISIS SENTIMEN PADA MEDIA SOSIAL X TERHADAP PROGRAM MAKAN SIANG GRATIS MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM)

Dwiputra, Bernards Bhima Immanuell August (2025) ANALISIS SENTIMEN PADA MEDIA SOSIAL X TERHADAP PROGRAM MAKAN SIANG GRATIS MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM). Bachelor thesis, Institut Teknologi Kalimantan.

[img] Text
11181017_cover.pdf
Restricted to Repository staff only until 4 October 2027.

Download (413kB) | Request a copy
[img] Text
11181017_statement_of_authenticity.pdf
Restricted to Repository staff only until 4 October 2027.

Download (378kB) | Request a copy
[img] Text
11181017_publishing_agreement.pdf
Restricted to Repository staff only until 4 October 2027.

Download (890kB) | Request a copy
[img] Text
11181017_approval_sheeet.pdf
Restricted to Repository staff only until 4 October 2027.

Download (373kB) | Request a copy
[img] Text
11181017_preface.pdf
Restricted to Repository staff only until 4 October 2027.

Download (862kB) | Request a copy
[img] Text
11181017_abstract_id.pdf
Restricted to Repository staff only until 4 October 2027.

Download (880kB) | Request a copy
[img] Text
11181017_abstract_en.pdf
Restricted to Repository staff only until 4 October 2027.

Download (880kB) | Request a copy
[img] Text
11181017_tables_of_content.pdf
Restricted to Repository staff only until 4 October 2027.

Download (930kB) | Request a copy
[img] Text
11181017_illustrations.pdf
Restricted to Repository staff only until 4 October 2027.

Download (923kB) | Request a copy
[img] Text
11181017_tables.pdf
Restricted to Repository staff only until 4 October 2027.

Download (917kB) | Request a copy
[img] Text
11181017_notations.pdf
Restricted to Repository staff only until 4 October 2027.

Download (918kB) | Request a copy
[img] Text
11181017_chapter_1.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_chapter_2.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_chapter_3.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_chapter_4.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_conclusions.pdf
Restricted to Repository staff only until 4 October 2027.

Download (925kB) | Request a copy
[img] Text
11181017_bibliography.pdf
Restricted to Repository staff only until 4 October 2027.

Download (918kB) | Request a copy
[img] Text
11181017_enclosure.pdf
Restricted to Repository staff only until 4 October 2027.

Download (898kB) | Request a copy
[img] Text
11181017_paper.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_presentation.pdf
Restricted to Repository staff only until 4 October 2027.

Download (1MB) | Request a copy
[img] Text
11181017_Form. TA-020.pdf
Restricted to Repository staff only until 4 October 2027.

Download (212kB) | Request a copy

Abstract

Program makan siang gratis adalah salah satu program yang dicetuskan oleh pemerintahan Prabowo-Gibran mulai akhir tahun 2024. Penelitian ini akan berfokus pada sentimen masyarakat dan performa algoritma Support Vector Machine (SVM) untuk pengolahan data teks. Data komentar berasal dari aplikasi X dan akan diklasifikasikan ke dalam dua kelas yaitu positif dan negatif. Evaluasi performa yang digunakan adalah Confusion matrix, accuracy, F1-Score, kurva ROC, dan Wordcloud dengan perbandingan komposisi data training dan data testing adalah 60:40, 70:30, dan 80:20. Nilai F1-Score tertinggi pada komposisi 80:20 sebesar 0,79. Sedangkan nilai F1-Score terrendah pada komposisi 70:30 sebesar 0,77. Nilai ROC tertinggi pada komposisi 60:40 dan terrendah pada komposisi 70:30. Hasil penelitian ini dihadapkan pada dua kondisi yaitu mesin yang melakukan pembelajaran leih banyak cenderung memperoleh hasil yang lebih tinggi, namun indikasi terjadi overfitting. Sedangkan mesin dengan data testing lebih banyak cenderung lebih stabil dalam melakukan analisis sentimen. Melalui visualisasi dan wordcloud dapat dianalisis bahwa sentimen masyarakat terhadap program ini masih cenderung negatif, pemerintah perlu mengadakan perbaikan di masa mendatang seperti aspek kebersihan, pengadaan dana, serta perkembangan dalam aspek sosial lainnya.

Item Type: Thesis (Bachelor)
Subjects: T Technology > T Technology (General)
Divisions: Jurusan Matematika dan Teknologi Informasi > Informatika
Depositing User: Bernards Bhima Immanuell August Dwiputra
Date Deposited: 15 Jul 2025 01:08
Last Modified: 15 Jul 2025 01:08
URI: http://repository.itk.ac.id/id/eprint/24576

Actions (login required)

View Item View Item