Presentation information

General Session

General Session » GS-10 AI application

[2F3-GS-10g] AI応用:テキスト処理

Wed. Jun 9, 2021 1:20 PM - 3:00 PM Room F (GS room 1)

座長:成松 宏美(NTT)

1:20 PM - 1:40 PM

[2F3-GS-10g-01] Automating Review Response Generation Using Review characteristics

〇Hisatoshi Igusa1, Fujio Toriumi1 (1. Tokyo University)

Keywords:review, response generation, RNN Encoder-Decoder model

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. Therefore, it is important for the hotel to respond appropriately to user reviews in order to acquire customers. However, responding to all reviews can be a burden on hotel employees, so support is required.In this research, we create a model that automatically generates hotel's responses from user reviews using RNN Encoder-Decoder models. Furthermore, we construct a model that incorporates the information of the user evaluations, which are the information accompanying the review, and the sentence length of the review reply. This improve the accuracy of review replies, and confirm that this information is useful in generating review replies.

Authentication for paper PDF access

A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.