Named entity identification method capable of combining attention mechanism and multi-target cooperative training

Abstract

The invention provides a named entity identification method capable of combining an attention mechanism and multi-target cooperative training. The method comprises the following steps that: (1) carrying out a preprocessing operation on training data, and through character hierarchy mapping, obtaining the character vector representation of a sentence; (2) inputting the character vector representation obtained in (1) into a bidirectional LSTM (Long Short Term Memory) network, and obtaining the character vector representation of each word; (3) through word hierarchy mapping, obtaining the word vector representation of each sentence; (4) through the attention mechanism, splitting the word vector representation obtained in (3) with the character vector representation obtained in (1), and transmitting into the bidirectional LSTM network to obtain the semantic characteristic vector of the sentence; and (5) aiming at the semantic characteristic vector obtained in (4), carrying out entity annotation on each word by a conditional random field, and decoding to obtain an entity tag.

Description

technical field [0001] The present invention relates to the field of neural network named entity recognition methods, and more specifically, relates to a named entity recognition method combined with an attention mechanism and multi-task cooperative training. Background technique [0002] Named Entity Recognition (NER), as the basic work of some complex tasks in the field of natural language processing (NLP) (such as information extraction, question answering system, machine translation), is to find relevant entities from a piece of natural language text, and Mark its location and type. As a research hotspot in the field of NLP, named entity recognition is a challenging task. On the one hand, it is usually difficult for people to obtain a large amount of labeled data for model training; on the other hand, the characteristics of entities are often ever-changing. Therefore, to obtain a model with strong generalization ability, a large number of features are often required. p...

AI Technical Summary

Problems solved by technology

Although these methods take into account the morphological characteristics of words at the character level, the splicing method is relatively simple and crude, and it is difficult to flexibly adjust the weight between character features and word features according to the training data; in addition, the specific training data distribution is not considered. The influence of weight, only through the backpropagation of the entity recognition task to adjust the weight of the character vector, it is difficult to learn the distribution law between characters

Images