源域数据增强与多兴趣细化迁移的跨域推荐模型

扫码查看

原文链接

NETL
NSTL
万方数据

中文摘要：针对跨域推荐任务中源域用户交互数据不丰富所导致的兴趣偏好建模困难问题，以及多个兴趣之间的关联被忽略问题，提出源域数据增强与多兴趣细化迁移的跨域推荐模型。该模型引入源域数据增强策略，为每个用户在源域中生成经过去噪处理的辅助序列，缓解用户在源域中的交互数据稀疏问题，获得更丰富的用户兴趣偏好。使用双序列多兴趣提取模块和多兴趣细化迁移模块，完成兴趣提取与多个兴趣的细化迁移。在基于 3 个公开跨域推荐评测任务的对比实验中，与最优的基线相比，提出方法的平均MAE降低了 22。86%，平均RMSE降低了19。65%，取得了最优的性能表现，证明了提出方法的有效性。

外文标题：Cross-domain recommendation model based on source domain data augmentation and multi-interest refinement transfer

外文摘要：A cross-domain recommendation model that utilizes source domain data augmentation and multi-interest refinement transfer was proposed in order to address the issues of difficulty in modeling interest preferences in cross-domain recommendation tasks caused by the lack of user interaction data in the source domain,as well as the problem of ignored associations between multiple interests.A source-domain data augmentation strategy was introduced,generating a denoised auxiliary sequence for each user in the source domain.Then the sparsity of user interaction data in the source domain was alleviated,and enriched user interest preferences were obtained.The interest extraction and multi-interest refinement transfer were implemented by utilizing the dual sequence multi-interest extraction module and the multi-interest refinement transfer module.Three publicly cross-domain recommendation evaluation tasks were conducted.The proposed model achieved the best performance compared with the best baseline,reducing the average MAE by 22.86%and the average RMSE by 19.65%,which verified the effectiveness of the method.

外文关键词：

cold-start problemcross-domain recommendationdata augmentationmulti-interest extractionmulti-interest refinement transfer

作者：

尹雅博、朱小飞、刘议丹

展开 >

作者单位：

重庆理工大学计算机科学与工程学院,重庆 400054

关键词：

冷启动问题跨域推荐数据增强多兴趣提取多兴趣细化迁移

基金：

国家自然科学基金资助项目重庆市自然科学基金资助项目重庆市教育委员会科学技术研究计划资助项目重庆理工大学研究生教育高质量发展行动计划资助项目

项目编号：

62141201CSTB2022NSCQ-MSX1672KJZD-M202201102gzlcx20233203

出版年：

2024

DOI：

10.3785/j.issn.1008-973X.2024.08.018

浙江大学学报(工学版)

浙江大学

浙江大学学报(工学版)

CSTPCD北大核心

影响因子：0.625

ISSN：1008-973X

年,卷(期)：2024.58(8)

参考文献量26