Keyword extraction from online product reviews based on bi-directional LSTM recurrent neural network

Yue WANG, J. ZHANG

Research output: Chapter in Book/Report/Conference proceedingChapters

20 Citations (Scopus)

Abstract

Online reviews are acknowledged as an important source of product information when customers make purchasing decisions. However, in the era of information overload, product review data on the Internet are too abundant and contain much irrelevant information. This makes it difficult for customers to find useful reviews. To solve this issue, some e-commerce websites provide keywords for product reviews, but these are generated beforehand and have the potential to distort customers' opinions of products. This paper presents an automatic keyword extraction method based on a bi-directional long short-memory (LSTM) recurrent neural network (RNN). The results of experiments conducted on product reviews obtain by data-crawling jd.com show that the proposed approach has a very high accuracy of keyword extraction. This can help reduce human annotation efforts in e-commerce. Copyright © 2017 IEEE.

Original languageEnglish
Title of host publicationProceedings of 2017 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2017
Place of PublicationUSA
PublisherIEEE
Pages2241-2245
ISBN (Electronic)9781538609484
DOIs
Publication statusPublished - 2017

Citation

Wang, Y., & Zhang, J. (2017). Keyword extraction from online product reviews based on bi-directional LSTM recurrent neural network. In Proceedings of 2017 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2017 (pp. 2241-2245). IEEE. https://doi.org/10.1109/IEEM.2017.8290290

Keywords

  • Deep learning
  • E-commerce
  • Product design

Fingerprint

Dive into the research topics of 'Keyword extraction from online product reviews based on bi-directional LSTM recurrent neural network'. Together they form a unique fingerprint.