Learning to rank products based on online product reviews using a hierarchical deep neural network

Ho Chang Lee, Hae-Chang Rim, Do Gil Lee

Research output: Contribution to journalArticle

Abstract

Product ranking based on online product reviews is a task of inferring relative user preferences between different products as a variant of entity-level sentiment analysis. Despite the complex relationship between the overall user's preference and individual diverse opinions, existing approaches generally employ empirical assumptions about sentiment features of the products of interest. In this paper, we propose a novel unified approach for learning to rank products based on online product reviews. Unlike existing approaches, it uses deep-learning techniques to extract the high-level latent review representation that contains the most semantic information in the learning process. For this approach, we extend the recently proposed hierarchical attention network to operate in the ranking domain. This network hierarchically learns optimal feature representations of the products and their reviews through the use of two-level attention-based encoders. To construct a more advanced ranking model, several features were added to give sufficient information about the relative user preferences, and two representative ranking loss functions, RankNet and ListNet, were applied. Furthermore, we demonstrate that this network outperforms the existing methods in sales rank prediction based on online product reviews.

Original languageEnglish
Article number100874
JournalElectronic Commerce Research and Applications
Volume36
DOIs
Publication statusPublished - 2019 Jul 1

Fingerprint

Sales
Semantics
Neural networks
Product review
Ranking
Learning to rank
Deep neural networks
User preferences
Deep learning
Learning process
Loss function
Sentiment analysis
Prediction
Sentiment

Keywords

  • Hierarchical deep neural network
  • Online product reviews
  • Product ranking

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Networks and Communications
  • Marketing
  • Management of Technology and Innovation

Cite this

Learning to rank products based on online product reviews using a hierarchical deep neural network. / Lee, Ho Chang; Rim, Hae-Chang; Lee, Do Gil.

In: Electronic Commerce Research and Applications, Vol. 36, 100874, 01.07.2019.

Research output: Contribution to journalArticle

@article{5e89b156d3a34d6580e388702f9497ca,
title = "Learning to rank products based on online product reviews using a hierarchical deep neural network",
abstract = "Product ranking based on online product reviews is a task of inferring relative user preferences between different products as a variant of entity-level sentiment analysis. Despite the complex relationship between the overall user's preference and individual diverse opinions, existing approaches generally employ empirical assumptions about sentiment features of the products of interest. In this paper, we propose a novel unified approach for learning to rank products based on online product reviews. Unlike existing approaches, it uses deep-learning techniques to extract the high-level latent review representation that contains the most semantic information in the learning process. For this approach, we extend the recently proposed hierarchical attention network to operate in the ranking domain. This network hierarchically learns optimal feature representations of the products and their reviews through the use of two-level attention-based encoders. To construct a more advanced ranking model, several features were added to give sufficient information about the relative user preferences, and two representative ranking loss functions, RankNet and ListNet, were applied. Furthermore, we demonstrate that this network outperforms the existing methods in sales rank prediction based on online product reviews.",
keywords = "Hierarchical deep neural network, Online product reviews, Product ranking",
author = "Lee, {Ho Chang} and Hae-Chang Rim and Lee, {Do Gil}",
year = "2019",
month = "7",
day = "1",
doi = "10.1016/j.elerap.2019.100874",
language = "English",
volume = "36",
journal = "Electronic Commerce Research and Applications",
issn = "1567-4223",
publisher = "Elsevier",

}

TY - JOUR

T1 - Learning to rank products based on online product reviews using a hierarchical deep neural network

AU - Lee, Ho Chang

AU - Rim, Hae-Chang

AU - Lee, Do Gil

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Product ranking based on online product reviews is a task of inferring relative user preferences between different products as a variant of entity-level sentiment analysis. Despite the complex relationship between the overall user's preference and individual diverse opinions, existing approaches generally employ empirical assumptions about sentiment features of the products of interest. In this paper, we propose a novel unified approach for learning to rank products based on online product reviews. Unlike existing approaches, it uses deep-learning techniques to extract the high-level latent review representation that contains the most semantic information in the learning process. For this approach, we extend the recently proposed hierarchical attention network to operate in the ranking domain. This network hierarchically learns optimal feature representations of the products and their reviews through the use of two-level attention-based encoders. To construct a more advanced ranking model, several features were added to give sufficient information about the relative user preferences, and two representative ranking loss functions, RankNet and ListNet, were applied. Furthermore, we demonstrate that this network outperforms the existing methods in sales rank prediction based on online product reviews.

AB - Product ranking based on online product reviews is a task of inferring relative user preferences between different products as a variant of entity-level sentiment analysis. Despite the complex relationship between the overall user's preference and individual diverse opinions, existing approaches generally employ empirical assumptions about sentiment features of the products of interest. In this paper, we propose a novel unified approach for learning to rank products based on online product reviews. Unlike existing approaches, it uses deep-learning techniques to extract the high-level latent review representation that contains the most semantic information in the learning process. For this approach, we extend the recently proposed hierarchical attention network to operate in the ranking domain. This network hierarchically learns optimal feature representations of the products and their reviews through the use of two-level attention-based encoders. To construct a more advanced ranking model, several features were added to give sufficient information about the relative user preferences, and two representative ranking loss functions, RankNet and ListNet, were applied. Furthermore, we demonstrate that this network outperforms the existing methods in sales rank prediction based on online product reviews.

KW - Hierarchical deep neural network

KW - Online product reviews

KW - Product ranking

UR - http://www.scopus.com/inward/record.url?scp=85068751968&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068751968&partnerID=8YFLogxK

U2 - 10.1016/j.elerap.2019.100874

DO - 10.1016/j.elerap.2019.100874

M3 - Article

AN - SCOPUS:85068751968

VL - 36

JO - Electronic Commerce Research and Applications

JF - Electronic Commerce Research and Applications

SN - 1567-4223

M1 - 100874

ER -