Aspect-Based Recommendation Model for Fashion Merchandising

Weiqing Li and Bugao Xu ()
Additional contact information
Weiqing Li: Xi’an Polytechnic University
Bugao Xu: University of North Texas

A chapter in Advances in Digital Marketing and eCommerce, 2021, pp 243-250 from Springer

Abstract: Abstract An aspect-based recommendation model (ARM) was proposed to detect local and global aspect representations in customer reviews available on ecommerce websites for fashion merchandising. This model was constructed with two independent paths to process user/item reviews simultaneously, and each path had a convolutional neural network (CNN), a long-short time memory network (LSTM) with attention mechanism to separately capture local aspect features and global aspect features. To enhance the generalization of the ARM model, the local and global aspect features from both user and item reviews were merged through mutual operations prior to the rating prediction. The Clothing, Shoes & Jewelry dataset from Amazon 5-core was used to train and test ARM. The significance of the extracted aspects regarding user preferences and item properties from the reviews were examined as opposed to several state-of-the-art fashion recommenders.

Keywords: Aspect-based recommendation; Review documents; Attention mechanism; Deep learning (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:prbchp:978-3-030-76520-0_25

Ordering information: This item can be ordered from
http://www.springer.com/9783030765200

DOI: 10.1007/978-3-030-76520-0_25

Access Statistics for this chapter

More chapters in Springer Proceedings in Business and Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().