EmEL-V: EL++ Ontology Embeddings for Many-to-Many Relationships

Tracking #: 3229-4443

This paper is currently under review
Biswesh Mohapatra
Arundhati Bhattacharya
Sumit Bhatia
Raghava Mutharaju
G. Srinivasaraghavan

Responsible editor: 
Guest Editors NeSy 2022

Submission type: 
Full Paper
Knowledge Graph (KG) embeddings provide a dense, low-dimensional representation of entities and relations in a Knowledge Graph and are used successfully for various applications such as reasoning, and missing link prediction, question answering and search. However, most of the existing KG embeddings only consider the network structure of the graph and ignore the semantics and the characteristics of the underlying ontology that provides crucial information about relationships between entities in the KG. Recent efforts in this direction involve learning embeddings for a description logic (logical underpinning for OWL 2 ontologies) named EL++. However, such methods consider all the relations defined in the ontology to be one-to-one which severely limits their performance and applications. We provide a simple and effective solution, named EmEL-V, to overcome this shortcoming that allows such methods to consider many-to-many relationships while learning embedding representations. Experiments conducted using three EL++ ontologies and one benchmark generated ontology on a reasoning task (class subsumption prediction) show substantial performance improvement over five baselines. Our proposed solution also paves the way for learning embedding representations for even more expressive description logics such as SROIQ. The source code and the instructions to run it are available at https://github.com/kracr/el-embeddings. The ontologies used in the evaluation are available at https://doi.org/10.5281/zenodo.7023568.
Full PDF Version: 
Under Review