SCI推荐系统论文整理（23~24年）KGCN、RippleNet、知识图谱、线上教育-CSDN博客

本文链接：https://blog.csdn.net/2201_75316947/article/details/142499809

知识图谱

1.The integration of knowledge graph convolution network with denoising autoencoder（知识图谱卷积网络与去噪自动编码器的集成）

作者：Gurinder Kaur, Fei Liu, Yi-Ping Phoebe Chen

关键字：Recommender system、Denoising autoencoder、Meta feature learning、Latent features representation

DOI:Redirectinghttps://doi.org/10.1016/j.engappai.2024.108792

摘要：

The knowledge graph convolution network (KGCN) is a recommendation model that provides a set of top recommendations based on knowledge graph developed between users, items, and their attributes. In this study, we integrate the KGCN model with denoising autoencoder (DAE) to improve its recommendation performance. A trained DAE is used to sample K-dimensional latent representation for each user, which then transforms that representation to generate a probability distribution over items. The relationship between acquired latent representation and the meta features is modelled using multivariate multiple regression (MMR) kernel. As a result, without the need for new configuration assessments, performance estimation of new data is pursued directly through MMR and the decoder of DAE. Empirically, we demonstrate that on real-world datasets, the proposed method substantially outperforms other state-of-the-art baselines. Movie-Lens 100K (ML-100K) and Movie-Lens 1M (ML-1M), two common MovieLens datasets, are used to verify the accuracy of the proposed approach. The results from experiments show significant improvement of 41.17% when the proposed method is applied on KGCN model. The proposed framework outperforms other state-of-the-art frameworks on Recall@K and normalized discounted cumulative gain (NDCG@K) metrics by achieving higher scores for Recall@5, Recall@10, NDCG@1, and NDCG@10.

知识图谱卷积网络（KGCN）是一种推荐模型，它根据在用户、项目及其属性之间开发的知识图谱提供一组热门推荐。在本研究中，我们将 KGCN 模型与去噪自动编码器（DAE）集成，以提高其推荐性能。使用经过训练的 DAE 对每个用户的 K 维潜在表示进行采样，然后转换该表示以生成项目的概率分布。获得的潜在表示与元特征之间的关系使用多元多元回归（MMR）内核进行建模。因此，无需新的配置评估，可以直接通过 MMR 和 DAE 的解码器对新数据进行性能估计。实证表明，在真实世界的数据集上，所提出的方法大大优于其他最先进的基线。Movie-Lens 100K （ML-100K）和 Movie-Lens 1M （ML-1M）是两个常见的 MovieLens 数据集，用于验证所提出的方法的准确性。实验结果表明，将所提方法应用于 KGCN 模型时，性能显著提高了 41.17%。拟议的框架在 Recall@K 和标准化折扣累积增益（NDCG@K）指标上优于其他最先进的框架，在 Recall@5、Recall@10、NDCG@1 和 NDCG@10 方面取得了更高的分数。

2.Multi-stream graph attention network for recommendation with knowledge graph（使用知识图谱进行推荐的多流图注意力网络）

作者：Zhifei Hu , Feng Xia

关键字：Graph attention network、Knowledge graph、Recommendation、Information propagation 【图注意力网络、知识图谱、推荐、信息传播】

DOI：

Redirectinghttps://doi.org/10.1016/j.websem.2024.100831

摘要：

In recent years, the powerful modeling ability of Graph Neural Networks (GNNs) has led to their widespread use in knowledge-aware recommender systems. However, existing GNN-based methods for information propagation among entities in knowledge graphs (KGs) may not efficiently filter out less informative entities. To address this challenge and improve the encoding of high-order structure information among many entities, we propose an end-to-end neural network-based method called Multi-stream Graph Attention Network (MSGAT). MSGAT explicitly discriminates the importance of entities from four critical perspectives and recursively propagates neighbor embeddings to refine the target node. Specifically, we use an attention mechanism from the user's perspective to distill the domain nodes' information of the predicted item in the KG, enhance the user's information on items, and generate the feature representation of the predicted item. We also propose a multi-stream attention mechanism to aggregate user history click item's neighborhood entity information in the KG and generate the user's feature representation. We conduct extensive experiments on three real datasets for movies, music, and books, and the empirical results demonstrate that MSGAT outperforms current state-of-the-art baselines.

近年来，图神经网络（GNN）强大的建模能力使其在知识感知推荐系统中的广泛应用。然而，现有基于GNN的方法在知识图谱（KG）中进行实体信息传播时，可能无法有效过滤掉不太重要的实体。为了解决这个挑战并提高多个实体之间高阶结构信息的编码，我们提出了一种端到端的神经网络方法，称为多流图注意力网络（MSGAT）。MSGAT从四个关键角度显式区分实体的重要性，并递归地传播邻居嵌入以精炼目标节点。具体而言，我们使用用户视角的注意力机制来提炼KG中预测项目的领域节点信息，增强用户对项目的信息，并生成预测项目的特征表示。我们还提出了一种多流注意力机制，以聚合KG中用户历史点击项目的邻域实体信息，并生成用户的特征表示。我们在三个真实数据集（电影、音乐和书籍）上进行了广泛的实验，实证结果表明，MSGAT的表现超过了当前最先进的基线。

3.Graph contrast learning for recommendation based on relational graph convolutional neural network（基于关系图卷积神经网络的图对比推荐学习）

作者：Xiaoyang Liu a, Hanwen Feng a, Xiaoqin Zhang b, Xia Zhou c, Asgarali Bouyer

关键字：Recommendation system、Graph convolutional neural network、Graph contrast learning、Knowledge grap 【推荐系统、图卷积神经网络、图谱对比学习、知识图谱】

DOI：Redirectinghttps://doi.org/10.1016/j.jksuci.2024.102168

摘要：

Current knowledge graph-based recommendation methods heavily rely on high-quality knowledge graphs, often falling short in effectively addressing issues such as the cold start problem and heterogeneous noise in user interactions. This leads to biases in user interest and popularity. To overcome these challenges, this paper introduces a novel recommendation approach termed Knowledge-enhanced Perceptive Graph Attention with Graph Contrastive Learning (KPA-GCL), which leverages relational graph convolutional neural networks. The proposed method optimizes the triplet embedding representation of entity-item interactions based on relationships between adjacent entities in a heterogeneous graph. Subsequently, a graph convolutional neural network is employed for enhanced aggregation. Similarity scores from a contrastive view serve as the selection criterion for high-quality embedded representations, facilitating the extraction of refined knowledge subgraphs. Multiple adaptive contrast-loss optimization functions are introduced by combining Bayesian Personalized Ranking (BPR) and hard negative sampling techniques. Comparative experiments are conducted with ten popular existing methods using real public datasets. Results indicate that the KPA-GCL method outperforms compared methods in all datasets based on Recall, NDCG, Precision, and Hit-ratio measures. Furthermore, in terms of mitigating cold start and noise, the KPA-GCL method surpasses other ten methods. This validates the reasonability and effectiveness of KPA-GCL in real-world datasets.

目前基于知识图谱的推荐方法在很大程度上依赖高质量的知识图谱，但往往无法有效解决冷启动问题和用户交互中的异构噪声，这导致用户兴趣和流行度的偏差。为了解决这些挑战，本文提出了一种新颖的推荐方法，称为知识增强感知图注意力与图对比学习（KPA-GCL），该方法利用关系图卷积神经网络。所提出的方法基于异构图中相邻实体之间的关系，优化实体-项目交互的三元组嵌入表示。随后，采用图卷积神经网络进行增强聚合。对比视角的相似性得分作为高质量嵌入表示的选择标准，促进精炼知识子图的提取。通过结合贝叶斯个性化排序（BPR）和困难负采样技术，引入多种自适应对比损失优化函数。使用十种流行的现有方法和真实公共数据集进行比较实验。结果表明，KPA-GCL方法在所有数据集的召回率、NDCG、精确度和命中率指标上均优于比较方法。此外，在缓解冷启动和噪声方面，KPA-GCL方法也超过了其他十种方法。这验证了KPA-GCL在真实数据集中的合理性和有效性。

图一阐述了 KPA-GCL 的综合框架，分为三个主要组成部分：知识感知嵌入、图卷积聚合和知识图谱感知对比学习。

图 2.在左图中，黄色节点代表中心用户节点，强调该用户的感受野。在右图中，蓝色节点代表用户节点，黄色三角形是项目节点，紫色区域表示当前中心用户节点的感受野的不同水平和范围。这表示特定用户可以影响的项或其他用户的集合。该图强调了用户感受野在推荐系统中的潜在间接影响。

4.Trustworthiness-awareknowledgegraphrepresentationfor recommendation（用于推荐的可信度感知知识图谱表示）

作者：Yan Ge a, Jun Ma b, Li Zhang c, Xiang Li a, Haiping Lu d

关键字：Recommender systems、Knowledge graph representation、Trustworthiness【推荐系统知识图谱表示可信度】

DOI:Redirectinghttps://doi.org/10.1016/j.knosys.2023.110865

摘要：

Incorporating knowledge graphs (KGs) into recommender systems (RS) has recently attracted increasing attention. For large-scale KGs, due to limited labour supervision, noises are inevitably introduced during automatic construction. However, the effects of such noises as untrustworthy information in KGs on RS are unclear, and how to retain RS performing well while encountering such untrustworthy information has yet to be solved. Motivated by them, we study the effects of the trustworthiness of the KG on RS and propose a novel method trustworthiness-aware knowledge graph representation (KGR) for recommendation (TrustRec). TrustRec introduces a trustworthiness estimator into noise-tolerant KGR methods for collaborative filtering. Specifically, to assign trustworthiness, we leverage internal structures of KGs from microscopic to macroscopic levels: motifs, communities and global information, to reflect the true degree of triple expression. Building on this estimator, we then propose trustworthiness integration to learn noise-tolerant KGR and item representations for RS. We conduct extensive experiments to show the superior performance of TrustRec over state-of-the-art recommendation methods.

将知识图谱（KG）纳入推荐系统（RS）近年来引起了越来越多的关注。对于大规模知识图谱，由于劳动监督有限，在自动构建过程中不可避免地引入了一些噪声。然而，这些作为不可信信息的噪声对推荐系统的影响尚不明确，而如何在面对这些不可信信息时保持推荐系统的良好性能仍待解决。基于此，我们研究了知识图谱的可信度对推荐系统的影响，并提出了一种新颖的方法，即基于可信度的知识图谱表示（KGR）用于推荐（TrustRec）。TrustRec在抗噪声的KGR方法中引入了可信度估计器，以便进行协同过滤。具体而言，为了分配可信度，我们利用知识图谱的内部结构，从微观到宏观层面：图案、社区和全局信息，来反