摘要
【目的】利用多模态特征融合和深度强化学习缓解数据稀疏性和用户兴趣偏好动态变化问题。【方法】使用预训练模型和注意力机制分别实现模态内表征和三模态间融合,同时建模用户-项目交互,并利用深度强化学习算法实时捕捉用户兴趣漂移和长短期奖励实现个性化推荐。【结果】较对比模型中最高值,所提模型在MovieLens-1M、MovieLens-100K和Douban数据集上的Precision@5分别提高11.8%、16.5%和11.4%,NDCG@5分别提高5.3%、8.0%和6.4%。【局限】Douban数据集的用户交互历史较少,所提模型在训练过程中无法学习到更准确的用户偏好,与在MovieLens数据集上的实验相比,推荐结果受限。【结论】所提模型融合项目多模态信息重构深度强化学习的状态表示网络,改善了推荐效果。
[Objective]This paper addresses data sparsity and dynamic changes in user interests with multimodal feature fusion and deep reinforcement learning.[Methods]First,we used a pre-trained model and attention mechanism to achieve intra-modal representation and fusion of three modalities.Then,we created a model for user-item interactions.Finally,we utilized the deep reinforcement learning algorithm to capture user interest drift and long and short-term rewards in real time to achieve personalized recommendations.[Results]Compared with the highest value in the baseline models,the proposed model improved precision@5 by 11.8%,16.5%,11.4%,and NDCG@5 by 5.3%,8.0%,6.4%,on the MovieLens-1M,MovieLens-100K,and Douban datasets,respectively.[Limitations]The user interaction history in the Douban dataset is relatively small,and the proposed model cannot learn more accurate user preferences during training.Compared with the experiments on the MovieLens dataset,we received limited recommendation results.[Conclusions]The proposed model integrates multimodal information to reconstruct the state representation network of deep reinforcement learning,improving the recommendation effect.
作者
潘华莉
谢珺
高婧
续欣莹
王长征
Pan Huali;Xie Jun;Gao Jing;Xu Xinying;Wang Changzheng(College of Information and Computer,Taiyuan University of Technology,Jinzhong 030600,China;College of Electrical and Power Engineering,Taiyuan University of Technology,Taiyuan 030024,China;Shanxi Tongfang Knowledge Network Digital Publishing Technology Co.,Ltd.,Taiyuan 030000,China)
出处
《数据分析与知识发现》
CSSCI
CSCD
北大核心
2023年第4期114-128,共15页
Data Analysis and Knowledge Discovery
基金
虚拟现实技术与系统国家重点实验室(北京航空航天大学)开放课题基金项目(项目编号:VRLAB2022C11)
山西省回国留学人员科研资助项目(项目编号:2020-040)
山西省科技合作交流专项项目(项目编号:202104041101030)的研究成果之一。
关键词
推荐
深度强化学习
多模态特征融合
用户-推荐系统交互
Recommendation
Deep Reinforcement Learning
Multimodal Feature
Fusion User-Recommender System Interaction
