A Heterogeneous Information Network Method for Entity Set Expansion in Knowledge Graph

Cao, Xiaohuan; Shi, Chuan; Zheng, Yuyan; Ding, Jiayu; Li, Xiaoli; Wu, Bin

doi:10.1007/978-3-319-93037-4_23

Cited by 8 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In KGs, the nodes represent entities and edges represent Resource Description Framework (RDF)like (<subject, property, object>) relations [14]. It is possible to represent a KG as a HIN, where the subject and object types are represented as a node and property as an edge [15]. As a result, KGs can be considered as an instance of HINs [14].…”

Section: A Heterogeneous Information Networkmentioning

confidence: 99%

“…However, KGs have some drawbacks for the computation due to their huge and complex structure. For example, in KGs enumeration of metapaths is impossible or the relations may indicate multiple meanings, i.e., the property of the edges can cause ambiguity [15]. In the literature, various methods, such as [16]- [23], use KGs for making recommendations.…”

Section: A Heterogeneous Information Networkmentioning

confidence: 99%

See 1 more Smart Citation

MP4Rec: Explainable and Accurate Top-N Recommendations in Heterogeneous Information Networks

et al. 2020

View full text Add to dashboard Cite

Neural network-based recommendation algorithms have become the state-of-the-art in recommender systems and can achieve very high predictive accuracy. However, these models are usually considered as black boxes in terms of their interpretability due to the complex structure of their hidden layers. In this research work, we propose MP4Rec, a recommender system using heterogeneous information networks to provide both accurate and explainable recommendations. MP4Rec uses of user-user and item-item similarity matrices and applies a newly proposed pair-wise objective function to make top-N recommendations which are transparent and explainable. The similarity matrices are created from metapaths constructed with the PathSim algorithm, node embeddings with cosine similarity or their combinations. The proposed pair-wise objective function incorporates an additional soft constraint for pushing more explainable items into the top-N recommendations. We have performed several experiments that show the effectiveness of our model by outperforming the state-of-the-art and providing both accurate and explainable recommendations in three well-known datasets.

show abstract

Section: A Heterogeneous Information Networkmentioning

confidence: 99%