NORLI, NORLI (2026) TOPIC- BASED CLUSTERING AND SOCIAL NETWORK ANALYSIS OF NEWS HEADLINES USING LDA AND DBSCAN. Project Report. UNIVERSITAS KATOLIK SOEGIJAPRANATA, Semarang. (Unpublished)
|
Text
22.K1.0054-NORLI-COVER _a.pdf Download (837kB) | Preview |
|
![[img]](http://repository.unika.ac.id/style/images/fileicons/text.png) |
Text
22.K1.0054-NORLI-ISI _a.pdf Restricted to Registered users only Download (1MB) |
|
|
Text
22.K1.0054-NORLI-COVER _a.pdf Download (837kB) | Preview |
|
|
Text
22.K1.0054-NORLI-LAMP _a.pdf Restricted to Registered users only Download (902kB) |
Abstract
Online news is growing rapidly every day, making it difficult for readers to find reading materials that suit their interests and understand issues currently being discussed in society. To address these issues, an automatic clustering method is needed to assist the information analysis process. The purpose of this study is to apply Latent Dirichlet Allocation (LDA) in extracting topic representations from news headlines and evaluate the effectiveness of the topic distribution in the clustering process using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method. The data used are a collection of news headlines from three online media, namely detik.com, kompas.com, and CNNIndonesia. The research stages include data pre-processing, text weighting, vector representation, application of LDA for topic formation, and application of DBSCAN to form clusters based on data density. Parameter determination is carried out through experiments to obtain the optimal number of topics in LDA as well as the appropriate epsilon and MinPts values in DBSCAN. The quality of the topic model is evaluated using the coherence score, while the quality of the cluster is assessed using the silhouette score. Furthermore, Social Network Analysis (SNA) was used to analyze the relationship structure between news headlines within each cluster through centrality and density measurements. The results indicated that the topic representation generated by LDA was effective as a basis for density-based clustering, and the network analysis was able to provide a deeper understanding of the relationship structure and the central role of news headlines within each cluster.
| Item Type: | Monograph (Project Report) |
|---|---|
| Subjects: | 000 Computer Science, Information and General Works > 004 Data processing & computer science |
| Divisions: | Faculty of Computer Science > Department of Informatics Engineering |
| Depositing User: | mr Dwi Purnomo |
| Date Deposited: | 11 Jun 2026 06:47 |
| Last Modified: | 11 Jun 2026 06:47 |
| URI: | http://repository.unika.ac.id/id/eprint/40000 |
| Keywords: | LDA, DBSCAN, SNA, preprocessing, online news clustering |
Actions (login required)
 |
View Item |