Pre-training model acquisition method and device, disease entity labeling method and device and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, in particular to the field of disease entity annotation of natural language processing, and discloses a pre-training model acquisition method and device, a disease entity labeling method and device and a storage medium, which can effectively improve the pre-training efficiency. The method comprises the steps of obtaining a first Chinese corpus, and performing word segmentation processing on the first Chinese corpus to obtain a preliminary word segmentation result; performing word segmentation on the non-common words of the preliminary word segmentation result to obtain a target word segmentation result; creating a Chinese vocabulary of a pre-training model according to the target word segmentation result; and pre-training the pre-training model by using the Chinese vocabulary to obtain a target pre-training model.

Description

technical field [0001] The present invention relates to the field of artificial intelligence technology, in particular to the application field of natural language processing for labeling disease entities, and in particular to a pre-training model acquisition and disease entity labeling method, device and storage medium. Background technique [0002] Migration learning is a very popular method in the field of deep learning. Accurate models can be established through migration learning, which takes less time. With transfer learning, instead of learning from scratch, start from the model learned when solving various problems before, avoiding training the model from scratch. [0003] Migration learning is usually represented by using pre-training models. The pre-training model is a model trained on a large benchmark data set, such as a large-scale pre-training language model such as BERT, which is obtained by pre-training a large amount of corpus. The proposal of the pre-train...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a vocabulary-based pre-training model acquisition method, a pre-training model-based disease entity labeling method, a device, and a storage medium to solve the problem that the pre-training model in the prior art is large in size and slow in reasoning, resulting in relatively low training efficiency. low problem

Images