Power dispatching field named entity identification method and system based on BiLSTM-CRF model

Abstract

The invention provides a power dispatching field named entity identification method and system based on a BiLSTM-CRF model, and the method comprises the steps: carrying out the word segmentation of anoriginal corpus, carrying out the labeling, and converting a labeling result into a first labeling result meeting the input format of the BiLSTM-CRF model; based on the BiLSTM-CRF model, identifyingthe first labeling result and outputting a second labeling result; and restoring the identified named entity of the power dispatching field from the obtained second labeling result. According to the named entity recognition method in the power dispatching field, the named entity recognition efficiency in the power dispatching field is improved to a certain extent, and it can be guaranteed that thenamed entity recognition result in the power dispatching field has good accuracy and recall rate.

Description

technical field [0001] The invention relates to the fields of electric power dispatching and computer technology, in particular to a named entity recognition method and system in the field of electric dispatching based on a BiLSTM-CRF model. Background technique [0002] With the continuous advancement of smart grid construction and the development of various smart grid businesses, China's smart grid has come to the forefront of the world in many aspects such as UHV, and the grid dispatching automation system is increasingly indispensable for the safe operation of smart grid. role. There are more and more various applications in the scheduling automation system, and the business knowledge is becoming more and more complex. Most of the relevant business personnel only understand partial business knowledge, but not the related business logic. Only a very small number of expert-level personnel can understand the entire business Process has a clear understanding. Therefore, wh...

AI Technical Summary

Problems solved by technology

[0004] However, relevant data show that the automatic recognition of named entities in the current power dispatching field is not ideal, and in most cases it is still necessary to rely on manual means for information extraction, especially when the text corpus is completely unstructured text, the rule-based extraction method is no longer Applicable, which makes it impossible for current staff to perform effective named entity recognition in unstructured power dispatch text corpus

Images