Power entity identification method in cloud environment

The power industry is accelerating the migration of business systems to the cloud. The knowledge graph in the cloud environment can describe the concepts, entities and relationships in massive data in a structured form, and has the ability to understand, organize and manage massive amounts of information more efficiently. The recognition of named entities is a basic technology. This paper proposes an entity recognition method. This method is based on a two-way long and short-term memory neural network, learns the long-term dependence information of the word vector sequence with the help of part of speech features, and then inputs a conditional random field to generate the global optimal sequence. Experiments show that this method is better than many traditional methods when performing named entity recognition on power corpus data. This method can not only lay a foundation for further construction of knowledge graphs and data mining systems, but also has certain reference significance for researchers and engineers in other fields.