Link Prediction
Are you familiar with #LinkPrediction?
The #KnowledgeGraph (KG) is a graph-based #data structure that represents a semantic network by nodes and edges. The edge represents the #relationship between two resources, while the node represents a resource (such as #concept and #individual).
For example, a #triple (Joe Russo, born_in, Cleveland) is represented as two entities: Joe Russo and Cleveland with a born_in relation that connects them. Knowledge graphs are used to describe concepts and the relations between them in the real world. Nowadays, knowledge graphs are widely used in the fields of #finance, #medical, #SemanticSearch, and other fields.
However, the KGs are still incomplete, i.e., missing a lot of valid triples. The goal of #KnowledgeGraphCompletion is to fill in the #MissingEdges (aka missing links) of a knowledge graph, i.e., edges that are deemed correct but aren’t given or implied by the knowledge graph. This task is often addressed with #Link_Prediction techniques.
Here, in the Borhan team, we work on a variety of knowledge graphs and try to predict missing links using #logical and #neural methods.