Claim Missing Document
Check
Articles

Found 1 Documents
Search

Sentimen Analisis & Topik Modeling menggunakan Latent Semantik Indexing (Studi Kasus: Aplikasi Detikcom) Glenzkovic Samuel Moniharapon; Dedi Iskandar Inan; Ratna Juita; Victor Arie Lambadya Sirait
G-Tech: Jurnal Teknologi Terapan Vol 8 No 4 (2024): G-Tech, Vol. 8 No. 4 Oktober 2024
Publisher : Universitas Islam Raden Rahmat, Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70609/gtech.v8i4.5434

Abstract

The proliferation of digital news platforms has transformed information consumption, making user experience (UX) a key success factor for Detik.com, a leading news application in Indonesia. This research aims to explore the relationship between UX and news through sentiment analysis and topic modelling. The methods include web scraping for data collection, descriptive analysis for sentiment, and Latent Semantic Indexing (LSI) for topic modelling. Out of 14,320 reviews analyzed, 45.6% were positive, focusing on the speed of updates and news quality. Conversely, 41.18% of reviews were negative, highlighting disruptive ads and login issues, while 13.76% were neutral. LSI identified eight main topics, including user experience and ad complaints. Visualizations through heatmaps, distance maps, and word clouds emphasized terms like "news" and "ads." This research recommends that Detik.com developers reduce ad disruptions, fix technical problems, and understand frequently discussed topics to enhance user satisfaction.