Power Domain Named Entity Recognition Based on Rules and Dictionaries

There are massive electricity data in the daily management, normalization operation, troubleshooting and other aspects of the power domain, but these professional and accurate data have not been fully mined and used. Constructing a power domain knowledge map can not only help power grid companies tap the value of these massive data and realize the integration of power knowledge, but also greatly facilitate the staff's query and acquisition of power information, and improve the work efficiency of the power industry. NER (name entity recognition) is the basis for constructing knowledge graph. This paper studies name entity recognition based on dictionaries and rules. It can standardize and accurately extract electricity from unstructured text through three methods: power entity dictionary, feature character rule matching, and part-of-speech combination rule matching. Related entities provide high-quality and high-precision entities for the construction of power domain knowledge graph.