Supervisión remota en el entrenamiento de un clasificador de sentimientos en comentarios turísticos
- Martín, C. 1
- Aguilar, R.M. 1
- Torres, J.M. 1
- Díaz S. 1
-
1
Universidad de La Laguna
info
- Inés Tejado Balsera (coord.)
- Emiliano Pérez Hernández (coord.)
- Antonio José Calderón Godoy (coord.)
- Isaías González Pérez (coord.)
- Pilar Merchán García (coord.)
- Jesús Lozano Rogado (coord.)
- Santiago Salamanca Miño (coord.)
- Blas M. Vinagre Jara (coord.)
Publisher: Universidad de Extremadura
ISBN: 978-84-9749-756-5, 978-84-09-04460-3
Year of publication: 2018
Pages: 644-650
Congress: Jornadas de Automática (39. 2018. Badajoz)
Type: Conference paper
Abstract
This paper describes an algorithm for automatically identifying the sentiments expressed by tourists on eWOM (Electronic Word of Mouth) platforms. Based on reviews published by tourists after staying in hotels, a classifier is trained using an LSTM (Long short-term memory supervised learning algorithm. The improvement of the training process with the use of remote supervision is demonstrated. The increase in the number of samples in the training set, although these present noise, improves the results of the classifier. We present a use case for this method involving a group of hotels located on the island of Tenerife (Canary Islands).