Title: Bi-LSTM-AMSM: bidirectional long short-term memory network and attention mechanism with semantic mining for e-commerce web page recommendation

Authors: Yachao Zhang; Yuxia Yuan

Addresses: School of Electronic and Electrical Engineering, Zhengzhou University of Science and Technology, China ' School of Electronic and Electrical Engineering, Zhengzhou University of Science and Technology, China

Abstract: Web page recommendation technology based on semantic mining has been widely used in e-commerce websites. When using the traditional web recommendation scheme, it cannot meet the user's application requirements. In this paper, we propose a novel model for e-commerce web page recommendation combing bidirectional long short-term memory network and attention mechanism with semantic mining. First, the web logs searched by users are processed to extract five features: content priority, time consumption priority, explicit/implicit feedback of e-commerce users for candidate sites, recommendation semantics and the amount of input bias. Then, these features are taken as input features of bidirectional long short-term memory network. Here, the bidirectional long short-term memory network is modified by attention mechanism to improve the classification effect. Third, the output priority of web page is identified by classification. Finally, the web pages are sorted by priority and recommended to users. Taking book sales web pages as samples, the results show that the bi-LSTM-AMSM scheme can quickly and accurately identify the web pages that users need.

Keywords: e-commerce; web page recommendation; semantic web mining; attention mechanism; bidirectional long short-term memory network.

DOI: 10.1504/IJWET.2023.136172

International Journal of Web Engineering and Technology, 2023 Vol.18 No.4, pp.376 - 396

Received: 30 Apr 2023
Accepted: 28 Oct 2023
Published online: 19 Jan 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article