This is the authors' abstract. We don't add key points for this paper.
The most critical step in the process of knowledge graph construction is entity relation extraction. However, most of the existing entity relationship extraction methods somewhat ignore common sense, resulting in poor results. In order to relieve the problem, in this paper, we propose a power triplet extraction method based on knowledge enhancement representation by combining text representation of domain and general knowledge enhancement. Firstly, the pre-training model is trained by using data sets in the power domain so that the pre-training model contains language features in the power domain. Then, through the graph embedding, the external general knowledge is embedded into the vector, and the output of the pre-trained model is stitched together so that the spliced vector contains both domain and general knowledge. Our method can relieve the problem of lack of common sense in the extraction of entity relationships. We conduct experiments on the power domain dataset and the results show that our model is better than the state-of-the-art baselines, and it has good performance in complex power application scenarios.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Zhu et al. (2022) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: