Learning Rule Embeddings over Knowledge Graphs: A Case Study from E Commerce Entity Alignment

E-Commerce platforms such as Alibaba and Amazon make people's life more convenient and play an important role in people's life. One critical issue for these platforms is entity alignment over item knowledge graph aiming at detecting if two items belong to the same product. In knowledge graph, rule learning methods and embedding-based methods are either inefficient or unexplainable. In this paper, we combine these two methods together and propose an embedding-based rule learning method that regard alignment rules as embeddings and learn these embeddings in the model training process. The experimental results on the real world E-Commerce dataset indicate that our proposed method is more effective and can reach better performance than other related methods.