%0 Journal Article
%T 基于二部图网络的电商平台智能化推送算法研究<br>Research on Intelligent Push Algorithm of E-Commerce Platform Based on Bipartite Graph Network
%A 李遇鑫
%A 谢本亮
%J E-Commerce Letters
%P 3067-3076
%@ 2168-5851
%D 2024
%I Hans Publishing
%R 10.12677/ecl.2024.132378
%X 在电商平台领域，大多数推荐算法都是基于用户–物品二部图网络(BGN)。但是这种推荐算法在准确性和多样性上严重不足。本文提出了一种基于BGN链接预测的电子商务推荐算法。首先，将所有用户项数据导入距离公式，计算属性之间的相似度；然后，将BGN投影到单模网络(SMN)中，提高了从BGN中提取潜在链路的效率。在此基础上，根据相似性对潜在链接进行预测。通过在真实电子商务数据集上的实验，证明了我们的算法比典型的推荐算法具有更高的准确率和覆盖率。<br />
In the field of e-commerce platform, most recommendation algorithms are based on user-item bipartite graph network (BGN). However, this recommendation algorithm is seriously insufficient in accuracy and diversity. In this paper, we propose an e-commerce recommendation algorithm based on BGN link prediction. Firstly, all user item data were imported into the distance formula to calculate the similarity between attributes. Then, the BGN was projected into a single mode network (SMN), which improved the efficiency of extracting potential links from the BGN. On this basis, the potential links are predicted based on the similarity. Through experiments on real e-commerce data sets, it is proved that our algorithm has higher accuracy and coverage than typical recommendation algorithms.
%K 二部图网络，机器学习，推送算法，电商平台智能化<br>Bipartite Graph Network
%K Machine Learning
%K Push Algorithm
%K Intelligent E-Commerce Platform
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=87953