• Publications
  • Influence
KGNN: Knowledge Graph Neural Network for Drug-Drug Interaction Prediction
This work proposes an end-to-end framework, called Knowledge Graph Neural Network (KGNN), to resolve the DDI prediction, which can effectively capture drug and its potential neighborhoods by mining their associated relations in KG. Expand
Repurpose Open Data to Discover Therapeutics for COVID-19 Using Deep Learning
This study built a comprehensive knowledge graph that includes 15 million edges across 39 types of relationships connecting drugs, diseases, proteins/genes, pathways, and expression from a large scientific corpus of 24 million PubMed publications and identified 41 repurposable drugs for COVID-19. Expand
Unsupervised Reference-Free Summary Quality Evaluation via Contrastive Learning
A new metric which covers both linguistic qualities and semantic informativeness based on BERT is designed which outperforms other metrics even without reference summaries and is general and transferable across datasets. Expand
Relation Matters in Sampling: A Scalable Multi-Relational Graph Neural Network for Drug-Drug Interaction Prediction
This work proposes an approach to modeling the importance of relation types for neighborhood sampling in graph neural networks and shows that it can learn the right balance: relation-type probabilities that reflect both frequency and importance. Expand
TLR4-mediated hippocampal MMP/TIMP imbalance contributes to the aggravation of perioperative neurocognitive disorder in db/db mice
It is suggested that TLR4-mediated aggravated hippocampal MMP/TIMP imbalance, BBB disruption, sustained inflammatory cytokine release, and impairment of long-term potentiation play a key role in tibial fracture surgery-induced persistent PND in db/db mice. Expand
Supervised Topic Compositional Neural Language Model for Clinical Narrative Understanding
  • X. Qin, Cao Xiao, +5 authors Fei Wang
  • Computer Science
  • IEEE International Conference on Big Data (Big…
  • 10 December 2020
A supervised topic compositional neural language model, called MeTRNN, that integrates the strength of supervised topic modeling in capturing global semantics with the capacity of contextual recurrent neural networks (RNN) in modeling local word dependencies. Expand
Biomimetic integrated olfactory sensory and olfactory bulb systems in vitro based on a chip.
A biomimetic integrated olfactory sensory and processing system can serve as a novel model for studying the physiological and pathological mechanisms of olfaction, and the pharmacological application in vitro. Expand
The graph convolutional networks (GCN) recently proposed by Kipf and Welling are an effective graph model for semi-supervised learning. This model, however, was originally designed to be learned withExpand