A TEXT PROCESSING METHOD AND DEVICE BASED ON AMBIGUOUS ENTITY WORDS

摘要

The present invention proposes a method and apparatus for processing text based on ambiguous entity words, wherein the text processing method based on ambiguous entity words is a context of text to remove ambiguity, and at least two candidate entities represented by text to remove ambiguity To obtain a semantic vector of the context through a trained word vector model, a first entity vector of at least two candidate entities, through a trained unsupervised neural network model, and a sync between the context and each candidate entity By calculating the rate, you determine the target entity that the text that you want to remove ambiguity expresses in context. The first entity vector of the generated candidate entity further includes the text meaning of the candidate entity and the relationship between each entity, through an unsupervised neural network model obtained by already learning the meaning of each entity text and the relationship between each entity, The entity information of the text to remove the ambiguity is completely shaped, and the semantic vector and sync rate of the context are calculated to determine the target entity, and the accuracy of removing the ambiguity is improved by determining the target entity.