Pengamatan Tren Ulasan Hotel Menggunakan Pemodelan Topik Berbasis Latent Dirichlet Allocation

Abstract

The accuracy in extracting and summarizing thousands of reviews into several topics is the key in the implementation of data processing and further information. The hotel industry is no exception, where a review is an asset which, when processed, can produce information that will later be used for business expansion and business continuity. This hotel review topic modeling research uses Latent Dirichlet Allocation as a means to summarize the document. Latent Dirichlet Allocation is proven to be effective in the processing of summarizing words and many studies have used this method. The purpose of this research is to get a summary of words that make up a topic that represents the whole review which can produce data for hotel management to maintain their existence in the business and expand by considering the results of modeling the topic. The results showed that the words location, service, hotel, breakfast, resort and beach were the terms that most often appeared among the dominant topics.