The knowledge graph representation learning is a key technology in the field of natural language process-ing.The existing research on knowledge graph representation mainly focuses on English and Chinese.This paper proposes a Joint Capsule Neural Network(JCapsR)model for Tibetan knowledge graph representation based on Ti-betan knowledge graph.Firstly,we use the TransR model to generate a structured information representation of the Tibetan knowledge graph.Secondly,we apply a a Transfomer model incorporating multiple-head attention and rela-tional attention to learn the entity text description information representation.Finally,we fuse the above two repre-sentations by the JCapsR model to obtain the final representation of Tibetan knowledge graph.The experimental re-sults show that the JCapsR model is more effective than the baselines in Tibetan knowledge graph representation learning.
knowledge graph of Tibetanrepresentation learningcapsule neural network
NETL
NSTL
万方数据
