We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
基于知识图谱嵌入的多跳中文知识问答方法.
- Authors
张天杭; 李婷婷; 张永刚
- Abstract
Based on the knowledge graph embedding model, a scoring method combining knowledge graph embedding scoring and link scoring is proposed to solve the multi-hop knowledge graph question answering task in the Chinese domain, which has wider applicability than the traditional single-hop knowledge question answering method. The method constructs a query link while searching for the optimal answer, and gives a set of answers through the query, thus effectively alleviating the situation of missing answers in the existing methods. The experimental results on the NLPCC-MH dataset show that the method is effective in The average F1 value on the multi - hop problem is 0.653, which is significantly better than the comparison method. The real knowledge graph usually has missing links. The experiment simulates the sparsity of the knowledge graph by randomly discarding 25% triples. The results show that The method still works in this case.
- Subjects
KNOWLEDGE graphs; TRADITIONAL knowledge
- Publication
Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban), 2022, Vol 60, Issue 1, p119
- ISSN
1671-5489
- Publication type
Article
- DOI
10.13413/j.cnki.jdxblxb.2020417