Abstract Graphs and Abstract Paths for Knowledge Graph Completion

Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM), pages 147–157 Minneapolis, June 6–7, 2019. c⃝2019 Association for Computational Linguistics 147 Abstract Graphs and Abstract Paths for Knowledge Graph Completion Vivi Nastase University of Heidelberg Heidelberg, Germany nastase@cl.uni-heidelberg.de Bhushan Kotnis NEC Laboratories Europe GmbH Heidelberg, Germany bhushan.kotnis@gmail.com Abstract Knowledge graphs, which provide numerous facts in a machine-friendly format, are incom- plete. Information that we induce from such graphs – e.g. entity embeddings, relation rep- resentations or patterns – will be affected by the imbalance in the information captured in the graph – by biasing representations, or caus- ing us to miss potential patterns. To par- tially compensate for this situation we describe a method for representing knowledge graphs that capture an intensional representation of the original extensional information. This rep- resentation is very compact, and it abstracts away from individual links, allowing us to find better path candidates, as shown by the results of link prediction using this information. 1 Introduction Knowledge graphs have become a very useful framework to organize and store knowledge. Their interconnected nature is not just a natural way to represent facts, but it has potential that the separate storage of facts does not have, such as: (i) we can use it as a relational model of meaning, and de- rive jointly representations for nodes (entities) and edges (relations); (ii) the structure can be explored to discover systematic patterns that reveal interest- ing and exploitable regularities, such as paths con- necting nodes in direct relations, (iii) discovering and inducing new connections. Link prediction methods in knowledge graphs (see (Nickel et al., 2016) for an overview) pre- dict additional edges in the graph, based on in- duced node and edge representations that encode the structure of the graph and thus capture regular- ities (such as homophily). Lao and Cohen (2010) introduced a new method that predicts direct links based on paths that con- nect the source and target nodes. Such paths are not only useful for link prediction (Lao et al., 2011; Gardner et al., 2014), but also for finding explanations for direct links and help with targeted information extraction to fill in incomplete knowl- edge repositories (Yin et al., 2018; Zhou and Nas- tase, 2018). These approaches rely on the structure of the knowledge graph, which is inherently incomplete. This incompleteness can affect the process in dif- ferent ways, e.g. it leads to representations for nodes with few connection that are not very in- formative, it can miss relevant patterns/paths (or derive misleading patterns/paths). In this paper we investigate whether a higher- level view of a graph – an abstract graph that cap- tures an intensional view of the original exten- sional graph – can help derive more robust and informative patterns. Such patterns are paths (i.e. sequences of relations) that could be used not only for link prediction, but also for targeted informa- tion extraction for completing the graph with ex- ternal information. This abstract graph will con- tain only one edge …

Content truncated to 500 words