Automatic Review Classification on Hotel Reservation Sites

Abstract

On the hotel reservation site, user reviews, evaluations and the hotel's response to them are extremely important information, and users make a hotel reservation with reference to these. However, responding to all reviews can be a burden on hotel employees, so support is required. In this research, we propose a method of automatic classification of reviews for creating automatic reply examples of reviews. A text classification model using convolutional neural network that considers the time series of words achieves a higher score than classification model using Bag of Words that doesn't consider the time series of words. In addition, it is possible to extract words that affect the results in estimating the evaluation.