Predicting missing and noisy links via neighbourhood preserving graph embeddings in a clinical knowlegebase.
Budhaditya SahaVedant MandloiShameek GhoshPublished in: ICMLA (2020)
Keyphrases
- missing data
- missing links
- link analysis
- graph representation
- incomplete data
- graph model
- graph theory
- graph structure
- strongly connected
- graph databases
- graph theoretic
- clinical practice
- directed graph
- random walk
- low dimensional
- patient records
- clinical trials
- clinical setting
- clinical data
- directed acyclic graph
- noisy data
- missing values
- patient data
- feature selection
- medical data
- noisy environments
- link graph
- vector space
- link formation