Identification method and device applied to electric power text named entity

Abstract

The invention relates to the technical field of electric power, in particular to an identification method and device applied to an electric power text named entity, and the method comprises the steps of inputting an open source electric power text obtained by a data collection module into a data preprocessing module, and obtaining preprocessed data; labeling the preprocessed data through data labeling software to obtain a labeled power text, and constructing a training set and a test set according to a preset proportion of the text of the labeled power text; inputting the training set into a joint learning model for training, predicting to obtain a tag sequence of the training set, and obtaining a named entity according to the tag; and testing the trained learning model by using the test set, and performing parameter optimization according to the accuracy rate and recall rate of the named entity recognition model, so as to solve the problems that a large amount of expensive tagged corpora are needed, noise data are mixed and the accuracy of the named entity recognition model is reduced when the named entity recognition model is applied to the power field in the prior art.

Description

technical field [0001] The present application relates to the field of electric power technology, in particular, to a method and device for identifying named entities in electric power texts. Background technique [0002] In recent years, my country's power technology has entered a period of rapid development, and power technology informatization and intelligence are also the development trend of the power industry. With the gradual deepening of power technology informatization, using computers to process and utilize large-scale power texts has become a hot spot in the field of power research. In text processing tasks, named entities take on a very important task. Named entities refer to noun objects with specific meanings in the text. Named entity recognition tasks aim to identify named entities in a given text. [0003] Due to the development and progress of artificial intelligence, people's demand for machines to process and understand text is becoming stronger and stron...

AI Technical Summary

Problems solved by technology

By combining reinforcement learning and deep learning models, it is possible to solve the problem of requiring a large amount of expensive annotation corpus, mixed with noisy data, and reducing the accuracy of named entity recognition models in the field of electric power.

Images