A Semantic Matching Energy Function for Learning with Multi-relational Data

TL;DR

This paper introduces a neural network architecture that embeds multi-relational graphs into a continuous space, capturing their semantics to improve link prediction in large-scale relational data.

Contribution

The paper proposes a novel semantic matching energy function and neural network model for embedding multi-relational data, enhancing the understanding of complex graph structures.

Findings

Achieves competitive link prediction performance on standard datasets

Effectively encodes semantics of multi-relational graphs

Demonstrates scalability to large datasets

Abstract

Large-scale relational learning becomes crucial for handling the huge amounts of structured data generated daily in many application domains ranging from computational biology or information retrieval, to natural language processing. In this paper, we present a new neural network architecture designed to embed multi-relational graphs into a flexible continuous vector space in which the original data is kept and enhanced. The network is trained to encode the semantics of these graphs in order to assign high probabilities to plausible components. We empirically show that it reaches competitive performance in link prediction on standard datasets from the literature.