<![CDATA[The vast amount of review data available on the Google Play Store can be utilized to extract hidden essential information. These reviews have an unstructured format that requiring particular methods to automatically collect and analyze the review data. Topic modeling is an extension of text analysis that can find main themes or trends hidden in large sets of unstructured documents. This study applies topic modeling with the Latent Dirichlet Allocation (LDA) method to Netflix application review data sourced from the Google Play Store web. The Latent Dirichlet Allocation (LDA) method is a generative probabilistic model from textual data that can explain the hidden semantic themes in the review document. This research aims to analyze hidden topics that application users discuss. These hidden topics contain essential valuable information for Netflix users and the company. Users can use this information to decide before using Netflix services. Meanwhile, Netflix can use this information to improve the quality of its services. This research use data from a web scraping Netflix review on the Google Play Store from January 2021–August 2021. The results of topic modeling show that of the twelve topics generated, the most discussed topic by users is payment methods.]]>
Copyrights © 2022
