KOMPARASI K-MEANS CLUSTERING DAN LATENT DIRICHLET ALLOCATION PADA KASUS SHORT TEXT TOPIC MODELING (STUDI KASUS : DATA JUDUL BERITA PEMILU 2024)

Enny Fitriani Nasution, NIM.: 19106050022 (2023) KOMPARASI K-MEANS CLUSTERING DAN LATENT DIRICHLET ALLOCATION PADA KASUS SHORT TEXT TOPIC MODELING (STUDI KASUS : DATA JUDUL BERITA PEMILU 2024). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (KOMPARASI K-MEANS CLUSTERING DAN LATENT DIRICHLET ALLOCATION PADA KASUS SHORT TEXT TOPIC MODELING (STUDI KASUS : DATA JUDUL BERITA PEMILU 2024)) 19106050022_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (KOMPARASI K-MEANS CLUSTERING DAN LATENT DIRICHLET ALLOCATION PADA KASUS SHORT TEXT TOPIC MODELING (STUDI KASUS : DATA JUDUL BERITA PEMILU 2024)) 19106050022_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

The 2024 Simultaneous Elections are an important democratic event for the nation of Indonesia. The role of the media as a source of information related to the elections is crucial in providing the fastest and most up-to-date information. Topic modeling is a method used to identify and analyze hidden patterns, themes, or topics within a collection of textual data. Therefore, this research aims to conduct topic modeling on news titles related to the 2024 elections using the K-Means Clustering and Latent Dirichlet Allocation (LDA) methods on online news portals. The objective of this research is to compare the two models and determine which of the two models performs better. The data used consists of news titles about the 2024 elections obtained through web scraping. Data analysis is performed using a confusion matrix and the Davies-Bouldin Index. In this research, the Latent Dirichlet Allocation (LDA) modeling was found to have an accuracy of 83.62%, precision of 86.29%, and recall of 83.62%. Meanwhile, K-Means Clustering has an accuracy 80.03%, precision 82.08%, and recall 80.00%. The DBI (Davies-Bouldin Index) for each model is 0.4979 for KMeans Clustering and 0.4085 for Latent Dirichlet Allocation (LDA).

Share this knowledge with your friends :

Actions (login required)

View Item