Representation Learning on Rdf and Lpg Knowledge Graphs

• hasURL https://towardsdatascience.com/representation-learning-on-rdf-and-lpg-knowledge-graphs-6a92f2660241
• hasTopic Labelled Property Graph
• hasHighlight
  • it’s important to notice that Wikidata (and generally the hyper-relational paradigm) does not separate between entities and predicates used exclusively in the main triples or qualifiers, i.e., "all predicates and entities can be used either in triple terms or qualifiers"
  • The field of graph representation learning (GRL) is one of the fastest-growing 🚀 areas of machine learning, there is a handful of articles (a series of posts by Michael Bronstein, reviews (mine, Sergey’s) from ICLR’20 and NeurIPS’19 papers), books (by William Hamilton, by Ma and Tang), courses (CS224W, COMP 766, ESE 680), and even a GraphML Telegram channel (subscribe 😉) covering basic and advanced topics.