桂林电子科技大学学报2025，Vol.45Issue(1)：11-19,9.DOI:10.16725/j.1673-808X.202322
使用交叉注意力融合项目类别属性的序列推荐模型
Sequence recommendation model with cross-attention fusion of item category attributes
摘要
Abstract
The changes in user preference in the interaction sequence can be reflected by the changes in item attributes in the se-quence.To further improve the accuracy of recommendation systems,item attributes can be integrated into sequence recommenda-tion algorithms in a reasonable way to achieve a more precise capturing of user interests.Previous methods used multi-layer percep-tron to fuse attribute information before the attention layer,which brought additional interference to the attention calculation due to the heterogeneity of item attributes and the correlation between item information and attribute information.The additional parame-ters brought by the multi-layer perceptron also increased the burden of the training process.This paper proposes selecting the project category that most intuitively reflects the changes in interest preferences as the attribute,separating the category interest extraction from the project prediction by decoupling the attribute representation from the project representation.The self-attention sparsifica-tion method is used to reduce the influence of noisy items in the sequence and extract more accurate category interests.Based on the extracted category interests,a multi-head cross-attention method is applied to aggregate the items in the sequence,avoiding the ag-gregation of items from different interest centers.Comparative and ablation experiments were conducted on real datasets such as Beauty,Sports,and Toys,and the results showed that the proposed model has better performance than baseline models and similar models.关键词
推荐算法/序列推荐/属性信息/多头交叉注意力方法/稀疏自注意力方法Key words
recommendation algorithm/sequence recommendation/attribute information/multi-head cross-attention method/sparse self-attention method分类
信息技术与安全科学引用本文复制引用
姜浩,蔡国永,湛永松,陈金龙..使用交叉注意力融合项目类别属性的序列推荐模型[J].桂林电子科技大学学报,2025,45(1):11-19,9.基金项目
广西科技重大专项(桂科AA19046004) （桂科AA19046004）
