Topic Modeling Using Latent Dirichlet Allocation (LDA) and Sentiment Analysis for Marketing Planning Tiket.com

DOI

10.2991/assehr.k.201010.004How to use a DOI?

Keywords

Tiketcom, Topic Modeling, Latent Dirichlet Allocation, Sentiment Analysis

Abstract

Tiket.com is a company that provides online ticket booking services in Indonesia. Tiketcom wants to improve services by knowing content that is widely discussed by the public and positive and negative comments on Tiketcom. Therefore an analysis will be done using Twitter accounts with the Latent Dirichlet Allocation (LDA) method which aims to find patterns in a document that raises various topics from text data and sentiment analysis to find out positive and negative comments on Tiketcom. The data used is tweet and retweet the users Twitter to Tiketcom accounts starting from 17 November 2018 to 4 March 2019. Obtained as many as 20 topics in the text data and taken 5 topics with the highest coherence value to obtain a topic model. After analyzing the LDA it was found that 5 topics that were widely discussed were promo discount tickets provided by Tiketcom. In sentiment analysis 21.1% of negative tweets were obtained, mostly discussing disruption to ticket reservations and 15.4% positive tweets mostly discussing vouchers given by Tiketcom to their customers.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Berlin Helmi Puspita
AU  - Muhammad Muhajir
AU  - Hafizhan Aliady
PY  - 2020
DA  - 2020/10/11
TI  - Topic Modeling Using Latent Dirichlet Allocation (LDA) and Sentiment Analysis for Marketing Planning Tiket.com
BT  - Proceedings of the 2nd International Seminar on Science and Technology (ISSTEC 2019)
PB  - Atlantis Press
SP  - 16
EP  - 22
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.201010.004
DO  - 10.2991/assehr.k.201010.004
ID  - Puspita2020
ER  -