An end-to-end neighborhood-based interaction model for knowledge-enhanced recommendation

Qu, Yanru; Bai, Ting; Zhang, Weinan; Nie, Jian‐Yun; Tang, Jian

doi:10.1145/3326937.3341257

Cited by 53 publications

(44 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, these technologies usually compress the information of a node and its neighborhood into single embedding vector before making prediction [4,22]. In this case, only two nodes and one edge are activated, yet other nodes and their connections are mixed and relayed, which introduces an early summarization issue [18]. The other school is metapath-based approaches.…”

Section: Introductionmentioning

confidence: 99%

An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph

Jin

Qin

et al. 2020

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

There is an influx of heterogeneous information network (HIN) based recommender systems in recent years since HIN is capable of characterizing complex graphs and contains rich semantics. Although the existing approaches have achieved performance improvement, while practical, they still face the following problems. On one hand, most existing HIN-based methods rely on explicit path reachability to leverage path-based semantic relatedness between users and items, e.g., metapath-based similarities. These methods are hard to use and integrate since path connections are sparse or noisy, and are often of different lengths. On the other hand, other graph-based methods aim to learn effective heterogeneous network representations by compressing node together with its neighborhood information into single embedding before prediction. This weakly coupled manner in modeling overlooks the rich interactions among nodes, which introduces an early summarization issue. In this paper, we propose an end-to-end Neighborhood-based Interaction Model for Recommendation (NIRec) to address above problems. Specifically, we first analyze the significance of learning interactions in HINs and then propose a novel formulation to capture the interactive patterns between each pair of nodes through their metapath-guided neighborhoods. Then, to explore complex interactions between metapaths and deal with the learning complexity on large-scale networks, we formulate interaction in a convolutional way and learn efficiently with fast Fourier transform. The extensive experiments on four different types of heterogeneous graphs demonstrate the performance gains of NIRec comparing with state-of-the-arts. To the best of our knowledge, this is the first work providing an efficient neighborhood-based interaction model in the HIN-based recommendations. CCS CONCEPTS • Information systems → Data mining; • Computer systems organization → Heterogeneous (hybrid) systems.

show abstract

Section: Introductionmentioning

confidence: 99%

An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph

Jin

Qin

et al. 2020

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…Finally, the user preference can be calculated viaŷ u,v = e * T u e * v . Qu et al [75] proposed KNI, which further considers the interaction between item-side neighbors and user-side neighbors. After obtaining high-order representation e (H) of entities in the KG, instead of using the enhanced user and item embedding to predict the user preference, KNI leverages the enhanced representation of user neighbors N u and item neighbors N v for preference estimation.…”

Section: Refinement Of Both User and Item Representationmentioning

confidence: 99%

A Survey on Knowledge Graph-Based Recommender Systems

Guo

Zhuang

Qin

et al. 2022

IEEE Trans. Knowl. Data Eng.

403

135

View full text Add to dashboard Cite

To solve the information explosion problem and enhance user experience in various online applications, recommender systems have been developed to model users' preferences. Although numerous efforts have been made toward more personalized recommendations, recommender systems still suffer from several challenges, such as data sparsity and cold-start problems. In recent years, generating recommendations with the knowledge graph as side information has attracted considerable interest. Such an approach can not only alleviate the above mentioned issues for a more accurate recommendation, but also provide explanations for recommended items. In this paper, we conduct a systematical survey of knowledge graph-based recommender systems. We collect recently published papers in this field, and group them into three categories, i.e., embedding-based methods, connection-based methods, and propagation-based methods. Also, we further subdivide each category according to the characteristics of these approaches. Moreover, we investigate the proposed algorithms by focusing on how the papers utilize the knowledge graph for accurate and explainable recommendation. Finally, we propose several potential research directions in this field.

show abstract

“…其中 p e;ej , p e;e 和 p ej ;ej 分别表示 e 与 e j , e 与 e 和 e j 与 e j 之间的路径, MovieLens-1M [10,12,13,18,19,44,46,47,50,63,74,91] Amazon e-commerce [66,67] MovieLens-20M [12,20,20,48 Figure 4 (Color online) An illustration of entity linking for news data 谱 [43] 进行匹配. 在电影数据中, 知识图谱上关于电影的相关补充特征 (例如: 导演、演员和电影类型等) 可以作为辅助信息有效地帮助我们提升推荐系统的效果.…”

Section: 基于知识图谱的推荐算法分类unclassified

A survey on knowledge graph-based recommender systems

et al. 2020

View full text Add to dashboard Cite

Figure 1 (Color online) An illustration of explainable knowledge graph-based recommendation 扮演着至关重要的角色, 用来促进商务业务发展以及辅助决策过程, 并且广泛地应用于各大电子商务 (淘宝、Amazon、eBay 等) 和多媒体 (MovieLens、豆瓣) 等网站中. 个性化推荐算法是推荐系统的核心, 其主要可以被分为 3 类, 即基于内容的推荐算法、基于协同过滤的推荐算法和混合推荐算法 [5]. 其中, 协同过滤方法通过利用用户历史的行为偏好数据构建模型, 取得了巨大的成功 [6∼8]. 这种方法的优势在于不需要类似基于内容的推荐算法那样对物品进行复杂的特征提取与建模. 尽管这类协同过滤方法通常是有效且普适的, 但其依旧存在着多种问题. 主要包括用户和物品之间的行为关系数据的稀疏问题和对新用户或者新物品进行推荐时存在的冷启动问题. 为此, 研究者尝试将协同过滤推荐算法和其他辅助信息相结合 (例如, 用户与物品的属性特征、用户社交网络信息等) 搭建混合推荐系统来解决以上问题, 从而提升推荐效果 [9∼11]. 知识图谱作为混合推荐系统中一类有效的辅助信息, 近年来受到了大量研究者的关注 [2, 10, 12∼15] .

show abstract

An end-to-end neighborhood-based interaction model for knowledge-enhanced recommendation

Cited by 53 publications

References 31 publications

An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph

An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph

A Survey on Knowledge Graph-Based Recommender Systems

A survey on knowledge graph-based recommender systems

Contact Info

Product

Resources

About