Recurrent Skipping Networks for Entity Alignment
This work addresses entity alignment in knowledge graphs, which is crucial for integrating heterogeneous data sources, but it appears incremental as it builds on existing embedding methods with a novel network architecture.
The paper tackles the problem of learning knowledge graph embeddings for entity alignment by addressing the inability of current methods to capture long-term dependencies, which leads to heavy reliance on prior alignments. The proposed RSN4EA method outperformed state-of-the-art embedding-based EA methods on real-world datasets and achieved comparable performance for knowledge graph completion.
We consider the problem of learning knowledge graph (KG) embeddings for entity alignment (EA). Current methods use the embedding models mainly focusing on triple-level learning, which lacks the ability of capturing long-term dependencies existing in KGs. Consequently, the embedding-based EA methods heavily rely on the amount of prior (known) alignment, due to the identity information in the prior alignment cannot be efficiently propagated from one KG to another. In this paper, we propose RSN4EA (recurrent skipping networks for EA), which leverages biased random walk sampling for generating long paths across KGs and models the paths with a novel recurrent skipping network (RSN). RSN integrates the conventional recurrent neural network (RNN) with residual learning and can largely improve the convergence speed and performance with only a few more parameters. We evaluated RSN4EA on a series of datasets constructed from real-world KGs. Our experimental results showed that it outperformed a number of state-of-the-art embedding-based EA methods and also achieved comparable performance for KG completion.