Research on the Application of Vocabulary Relation Extractio n Method of Demand Entity Based on Bi-GRU
Jiakun Zhao · Journal of Physics Conference Series · 2021
Abstract Designing and mining the entity vocabulary relationship of requirements engineering to form a knowledge base of requirements engineering has a positive effect on providing the intelligent service of demand and software development in the future. But there are very few targeted extractions of the relationship between demand vocabulary in the current demand field. Therefore, this paper proposes a demand entity vocabulary relationship extraction model consists of the Bi-GRU, the attention mechanism and the labeling and filtering layer of the demand entity. This paper discribes how to use the data from actual projects to verify the methods above. The experimental results show that the method of relationship extraction combining the attention mechanism and the labeling and filtering layer of the demand entity in the Bi-GRU neural network has a significant effect in the demand vocabulary relationship extraction.