An Entity Alignment Algorithm Based on GCN and Translation Model
Jiangwei Tian, Qing Liu · 2020
Entity alignment (EA) is the core technology of building large-scale knowledge base and realizing knowledge fusion. In recent two years, many researches use the graph convolution neural network (GCN) for entity alignment and have achieved good results. However, the existing GCN-based EA algorithms do not effectively utilize the semantic information between entities. In this paper, an EA method combining GCN with translation model is proposed. It separately learns the embedded vectors of entities based on GCN and translation model, and then computes the distance of vectors to align entities. Experiments on real-world datasets show the effectiveness of the proposed approach.