Opinion mining on mandalika hotel reviews using latent dirichlet allocation

Research output: Contribution to journalConference articlepeer-review

42 Citations (Scopus)

Abstract

Mandalika in the Province of West Nusa Tenggara (NTB) is one of the top 10 priority tourist destinations in Indonesia. Mandalika is also a Tourism Special Economic Zone because of its great tourism potential. The Special Tourism Economic Zone is an area designated for tourism business activities to support the implementation of entertainment and recreation, meetings and related activities. It is supported by the provision of infrastructure and ease of investment. The development of Mandalika is still being carried out by all parties to make continuous quality improvements, including hotel management. Hotel as one of tourism supporting factors is located around the tourism destinations and it provides facilities and services for tourists. Therefore, hotel can be one of traveler experience subjects which is often posted and discussed in online media. Hotel management can understand whether its business is doing well and what they need to improve. This study aims to infer the topics which is extracted from hotel reviews using Latent Dirichlet Allocation. The output of this study is eight topics extracted using topic coherence as the evaluation measurement for topic modeling using LDA. These topics are often discussed which certainly can be a feedback between hotel management and tourists in order to increase a greater number of tourists visiting to Mandalika.

Original languageEnglish
Pages (from-to)739-746
Number of pages8
JournalProcedia Computer Science
Volume161
DOIs
Publication statusPublished - 1 Jan 2019
Event5th Information Systems International Conference, ISICO 2019 - Surabaya, Indonesia
Duration: 23 Jul 201924 Jul 2019

Keywords

  • Hotel reviews
  • Latent dirichlet allocation
  • Mandalika
  • Opinion mining
  • Tourism

Fingerprint

Dive into the research topics of 'Opinion mining on mandalika hotel reviews using latent dirichlet allocation'. Together they form a unique fingerprint.

Cite this