Named entity recognition method based on neural network model

• Main Authors: QIU SHAOMING, QIN XUTONG, DU XIULI, LYU YANA
• Format: Patent
• Language: Chinese; English
• Published: 31.01.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a named entity recognition method based on a neural network model. The named entity recognition method comprises the following steps: constructing a named entity recognition model for recognizing entities in a Chinese text; constructing a data set for training a named entity recognition model, and dividing the data set into a training set and a test set; based on a training set and a test set, training the named entity recognition model to obtain a trained named entity recognition model; processing the Chinese text needing named entity recognition into a specified format; a Chinese text processed into a specified format is input into a trained Chinese-oriented named entity recognition model, named entities in the Chinese text are recognized, a CNNs layer of the model has convolution kernels of various sizes, boundary information of all candidate entities can be encoded while word-level information is extracted, and the named entities in the model are recognized. The problem that the NE