Jurnal Gaussian
Vol 15, No 1 (2026): Jurnal Gaussian

LATENT DIRICHLET ALLOCATION DALAM IDENTIFIKASI RESPON MASYARAKAT INDONESIA TERHADAP PROFESI PEGAWAI NEGERI SIPIL

Aghentika, Nurul Fajrin (Unknown)
Sugito, Sugito (Unknown)
Warsito, Budi (Unknown)



Article Info

Publish Date
30 Apr 2026

Abstract

Data on social media comments can be extracted to produce hidden information that is useful as a guide for evaluation and decision making. YouTube has a comments feature as a forum for expressing opinions, experiences and questions. Civil Servants are known as one of the job choices of Indonesian people, the government announced that there were resignations of Civil Servant Candidates in 2022. Responses written in the comment column are difficult to understand, topic modeling can be applied as a text analysis process to find descriptions from unstructured data. Latent Dirichlet Allocation method is able to find out hidden topics in a document as well as the words that make up a topic so that the application of this method will help in identifying responses discussed by the audience. The data used is textual data in the form of comments from YouTube scrapping during 2022. The results of topic modeling form eight topics, namely retirement life, parents hopes, dream jobs, civil servants, job differences, characteristics of generation Z, salary and benefits, and reasons for resignation. The RStudio GUI program can make it easier for users to analyze topic modeling with similar methods.

Copyrights © 2026






Journal Info

Abbrev

gaussian

Publisher

Subject

Other

Description

Jurnal Gaussian terbit 4 (empat) kali dalam setahun setiap kali periode wisuda. Jurnal ini memuat tulisan ilmiah tentang hasil-hasil penelitian, kajian ilmiah, analisis dan pemecahan permasalahan yang berkaitan dengan Statistika yang berasal dari skripsi mahasiswa S1 Departemen Statistika FSM ...