3 Hits in 2.7 sec

MultiImport: Inferring Node Importance in a Knowledge Graph from Multiple Input Signals [article]

Namyong Park, Andrey Kan, Xin Luna Dong, Tong Zhao, Christos Faloutsos
2020 pre-print
Given multiple input signals, how can we infer node importance in a knowledge graph (KG)?  ...  In this paper, we develop an end-to-end model MultiImport, which infers latent node importance from multiple, potentially overlapping, input signals.  ...  We formulate the problem of inferring node importance in a KG from multiple input signals. • Algorithm.  ... 
doi:10.1145/3394486.3403093 arXiv:2006.12001v1 fatcat:tcmlgpa57vdltjfda4bmt3b4ee

J-Recs: Principled and Scalable Recommendation Justification [article]

Namyong Park, Andrey Kan, Christos Faloutsos, Xin Luna Dong
2020 arXiv   pre-print
The challenge of jointly processing multiple types of data is addressed by designing a principled graph-based approach for justification generation.  ...  Existing post-hoc methods are often limited in providing diverse justifications, as they either use only one of many available types of input data, or rely on the predefined templates.  ...  GENI [38] and MultiImport [39] are semi-supervised techniques to estimate node importance by considering both the graph structure and real-world signals of node popularity.  ... 
arXiv:2011.05928v1 fatcat:3dqm6aqazzbc5fcmwiapv7ewza

Mining and Learning With Graphs and Tensors

Namyong Park
We first propose novel graph-regularized semi-supervised algorithms for estimating node importance in a knowledge graph, which achieve up to 25% higher accuracy than the best baseline.Then we develop distributed  ...  In addition, we develop a method that explains product recommendations, up to 21% more accurately than the best baseline, by performing personalized inference over a product graph.  ...  Chapter 3 Inferring Node Importance in a Knowledge Graph from Multiple Input Signals 2 Chapter based on work published in KDD 2020 [PKD + 20]. Definition 3.  ... 
doi:10.1184/r1/19891765.v1 fatcat:w47xj5l3snehjkrdmp77yqphve