Semantic and Structural View Fusion Modeling for Social Recommendation
Semantic and Structural View Fusion Modeling for Social Recommendation
复制标题
用于社交推荐的语义和结构视图融合建模
DOI：
10.1109/tkde.2022.3230972
复制
发表时间：
2023-11
影响因子：
8.9
通讯作者：
Hui Xiong
中科院分区：
文献类型：
--
作者：
Kun Yuan;Guannan Liu;Junjie Wu;Hui Xiong
已有的研究表明，用户-项目交互数据和社会关系数据可以联合使用，以提高社会推荐的性能。然而，有限的研究集中在研究如何深入利用社会互动结构和评级行为差异的不同观点，以进一步提高社会推荐。为此，在本文中，我们建议整合信息从语义和结构的意见，社会推荐。具体来说，我们首先设计了一个基于集体智慧的策略，以揭示用户和项目的高质量的隐式关系。然后，通过重新制定所有可用的节点和关系作为一个异构图，我们定义了多个语义元路径，以捕捉不同的偏好，全面的用户和项目表示。各种元路径在扩大用户和项目的表征能力的同时，也引入了噪声和无关信息。我们记得，对于用户-项目交互图，不同的结构大小（例如，地方和全球结构）提供各种各样的补充信息供推荐。基于此，本文提出了一个语义和结构化视图融合的社会推荐框架（S4 Rec），该框架由深度图模型和广义奇异值分解（SVD）模型组成，通过将局部和全局结构作为输入，沿着沿着预定义的元路径聚合消息，进行评分预测。最后，对两种预测结果进行自适应融合，实现最终的既准确又稳定的预测。此外，我们将用户的评分行为差异视为嵌入空间中的相对位置差异问题，用transH对其进行建模，以提高主评分模型的泛化能力。在三个开放数据集上进行的大量实验表明，与最先进的方法相比，我们的框架具有优越性。特别是，我们的模型在不同稀疏条件下优于其他基线，进一步验证了冷启动用户的有效性。我们在https://github.com/lcwy220/Social-Recommendation上发布源代码。
Existing studies have shown that user-item interaction data and social relation data can be jointly used for enhancing the performance of social recommendation. However, limited research has a focus on investigating how to deeply exploit different views of social interaction structures and rating behavior differences for further improving social recommendation. To this end, in this paper, we propose to integrate information from both semantic and structural views for social recommendation. Specifically, we first design a collective intelligence-based strategy to reveal high-quality implicit relations for both users and items. Then, by reformulating all available nodes and relations as a heterogeneous graph, we define multiple semantic metapaths to capture diverse preferences for comprehensive user and item representations. While various metapaths enlarge the representation capacity of users and items, they also introduce noise and irrelevant information. We recall that, for the user-item interaction graph, different structure sizes (e.g., local and global structures) provide diverse and complementary information for recommendation. Motivated by this, we propose a semantic and structural view fusion framework for social recommendation (S4Rec), which consists of a deep graph model and a wide attentive SVD (Singular Value Decomposition) model for rating prediction by taking the local and global structure as input and aggregating messages along the predefined metapaths. Finally, the two predicted results are adaptively fused to achieve the final both accurate and stable prediction. In addition, we treat the user's rating behavior difference as the relative position difference problem in the embedding space, and model it with TransH to improve the generalization ability of the main rating model. Extensive experiments on three open datasets demonstrate the superiority of our framework compared with state-of-the-art methods. Particularly, our model outperforms other baselines under different sparsity conditions, further validating the effectiveness on cold-start users. We release the source code at https://github.com/lcwy220/Social-Recommendation.
登录
查看更多内容
DOI：
10.1109/tkde.2020.2982878
发表时间：
2020-03
影响因子：
8.9
作者：
Hongxu Chen;Hongzhi Yin;Tong Chen;Weiqing Wang;Xue Li;Xia Hu
通讯作者：
Hongxu Chen;Hongzhi Yin;Tong Chen;Weiqing Wang;Xue Li;Xia Hu
DOI：
10.1145/1557019.1557067
发表时间：
2009-06
期刊：
--
影响因子：
--
作者：
Mohsen Jamali;M. Ester
通讯作者：
Mohsen Jamali;M. Ester
DOI：
10.1145/3357384.3357898
发表时间：
2019-11
期刊：
Proceedings of the 28th ACM International Conference on Information and Knowledge Management
影响因子：
--
作者：
A. Krishnan;Hari Cheruvu;Tao Cheng;H. Sundaram
通讯作者：
A. Krishnan;Hari Cheruvu;Tao Cheng;H. Sundaram
DOI：
10.1609/aaai.v28i1.8714
发表时间：
2014-06
期刊：
--
影响因子：
--
作者：
Hui Fang;Yang Bao;Jie Zhang
通讯作者：
Hui Fang;Yang Bao;Jie Zhang
DOI：
10.1137/1.9781611974973.43
发表时间：
2017
期刊：
--
影响因子：
--
作者：
Chuxu Zhang;Lu Yu;Yan Wang;C. Shah;Xiangliang Zhang
通讯作者：
Chuxu Zhang;Lu Yu;Yan Wang;C. Shah;Xiangliang Zhang