TOPIC ANALYSIS OF TIMOR LESTE NEWS IN INDONESIA'S MASS MEDIA USING LATENT DIRICHLET ALLOCATION
DOI:
https://doi.org/10.25078/ijoss.v2i1.3916Keywords:
Timor Leste, news, Latent Dirichlet Allocation, Detik.comAbstract
This study analyzes news about Timor Leste in the Indonesian national mass media Detik.com using the Latent Dirichlet Allocation (LDA) method. This research is important to understand how the national mass media in Indonesia, especially Detik.com, report on Timor Leste. Timor-Leste is a neighboring country with a significant history and diplomatic relations with Indonesia. There is a gap between the existing literature, which often focuses on the political and historical links between the two countries, and the reality of everyday reporting covering various aspects of life in Timor-Leste. The aim of this study was to identify and analyse key topics in Timor-Leste reporting in Detik.com, in order to provide a more comprehensive picture of how Timor-Leste was presented to the Indonesian public. The methods used in this study include text mining, text preprocessing, and Latent Dirichlet Allocation (LDA). Using the LDA, we identified two main topics that emerged from articles discussing Timor Leste. The first topic (30.3% of tokens) focused on Timor-Leste's membership in ASEAN and relations with Indonesia, with keywords such as "asean", "leste", "timor", "year", "member", and "indonesia". The second topic (14.7% of tokens) revolved around local and sporting issues in Timor-Leste, with keywords such as "timor", "indonesia", "leste", "dili", "language", "minutes", "year", "goals", "east", and "second". The other eight topics deal with similar terms. The results showed that news about Timor-Leste in Detik.com covered aspects of politics and international relations as well as local developments and sports in the country. This research provides insight into how national media construct and convey information about Timor-Leste to the Indonesian public.