DEA ANGGRAENI, . (2025) PENERAPAN RANDOM FOREST DENGAN ALGORITMA LEVENSHTEIN DISTANCE UNTUK ANALISIS SENTIMEN TERHADAP ISU KEBIJAKAN BARU PROGRAM BEASISWA LPDP. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.
|
Text
COVER.pdf Download (2MB) |
|
|
Text
BAB I.pdf Download (522kB) |
|
|
Text
BAB II.pdf Restricted to Registered users only Download (1MB) | Request a copy |
|
|
Text
BAB III.pdf Restricted to Registered users only Download (674kB) | Request a copy |
|
|
Text
BAB IV.pdf Restricted to Registered users only Download (1MB) | Request a copy |
|
|
Text
BAB V.pdf Restricted to Registered users only Download (505kB) | Request a copy |
|
|
Text
DAFTAR PUSTAKA.pdf Download (526kB) |
|
|
Text
LAMPIRAN.pdf Restricted to Registered users only Download (2MB) | Request a copy |
Abstract
Media sosial menjadi wadah utama masyarakat dalam menyampaikan opini terhadap berbagai isu publik, termasuk kebijakan pemerintah. Kebijakan baru Program Beasiswa LPDP tahun 2024, yang memberikan kelonggaran bagi penerima untuk tidak wajib kembali ke Indonesia setelah studi di luar negeri, memunculkan beragam sentimen di platform X (Twitter). Penelitian ini menganalisis sentimen publik terkait kebijakan tersebut dengan menerapkan algoritma Levenshtein Distance pada tahap normalisasi teks dan Random Forest sebagai model klasifikasi. Data dikumpulkan melalui web scraping terhadap unggahan berbahasa Indonesia di platform X periode 31 Oktober 2024 hingga 28 Februari 2025 menggunakan kata kunci 'LPDP'. Hasil evaluasi model pada tiga skenario menunjukkan peningkatan akurasi dari 83,58% pada baseline menjadi 84,05% pada model dengan normalisasi Levenshtein Distance murni, dan mencapai 86,82% pada model dengan normalisasi Levenshtein Distance berbasis filtering rules dan scoring. Analisis sentimen menggunakan model terbaik menghasilkan distribusi 26,7% positif, 34,3% negatif, dan 39,0% netral yang mencerminkan keberagaman persepsi masyarakat sekaligus memberikan gambaran komprehensif mengenai opini publik terhadap kebijakan baru LPDP. ***** Social media has become a primary platform for the public to express opinions on various public issues, including government policies. The new 2024 LPDP Scholarship Program, which allows recipients not to return to Indonesia after studying abroad, has generated diverse sentiments on platform X (Twitter). This study analyzes public sentiment regarding the policy by applying the Levenshtein Distance algorithm for text normalization and Random Forest as the classification model. Data were collected through web scraping of Indonesian-language posts containing the keyword 'LPDP' from October 31, 2024, to February 28, 2025. Model evaluation across three scenarios showed an accuracy increase from 83.58% in the baseline scenario to 84.05% with pure Levenshtein Distance normalization, and up to 86.82% with normalization based on filtering rules and scoring. Sentiment analysis using the best performing model revealed a distribution of 26.7% positive, 34.3% negative, and 39.0% neutral, reflecting the diversity of public perception while providing a comprehensive overview of public opinion on the new LPDP policy.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Additional Information: | 1). Devi Eka Wardani Meganingtyas, S.Pd., M.Si. ; 2). Qorry Meidianingsih, M.Si. |
| Subjects: | Sains > Matematika |
| Divisions: | FMIPA > S1 Matematika |
| Depositing User: | Users 32818 not found. |
| Date Deposited: | 05 Feb 2026 03:51 |
| Last Modified: | 05 Feb 2026 03:51 |
| URI: | http://repository.unj.ac.id/id/eprint/64874 |
Actions (login required)
![]() |
View Item |
