A word vector embedding method and device

Abstract

The embodiment of the invention provides a word vector embedding method and device. The method comprises the steps of obtaining a reference word vector of each word matched with a pre-trained word vector library in a sentence where a target word is located; Determining an initial word vector of the target word according to the reference word vector; And training a target corpus set corresponding to the target word according to the initial word vector and a vector embedding model W2V, and determining an embedded word vector of the target word. According to the word vector embedding method, in the initialization stage, the target words are endowed with priori knowledge; Even if the corpus set is small or the target word does not appear in the pre-training corpus set, a reasonable target wordvector can be trained through the vector embedding model, so that the embedding vector of the target word is closer to the real semantics of the target word, the problem that the word vector embedding of the small corpus set is unreasonable is solved, the word vector embedding accuracy is improved, and the efficiency of the vector embedding model is further improved.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of natural language processing, and in particular to a word vector embedding method and device. Background technique [0002] Vector embedding (Word2Vec, W2V) is a Natural Language Processing (NLP) method. W2V vectorizes all the words in the text, so that it can quantitatively measure the relationship between words and mine the relationship between words. contact. The current general-purpose vector embedding tools mainly include the Continuous Bag-of-Words Model (CBOW) and the Skip-Gram model. The training input of the CBOW model is the word vector corresponding to the context-related word of the target word, and the word vector of the target word is output. The Skip-Gram model is the opposite of CBOW. It assumes that similar words have similar contexts. According to the context of the current word prediction, the input is a word vector of the target word, and the output is the contex...

AI Technical Summary

Problems solved by technology

[0007] However, for small-scale companies or remote fields, the corpus is small, and the target word appears less frequently in the corpus, or even does not appear in the corpus. Using Skip-Gram to randomly initialize word vectors for training leads to unreasonable word vector embedding, which affects performance of the Skip-Gram model

Images