Method and device using wikipedia link structure to generate chinese language concept vector

摘要

The present invention provides a method and device using a Wikipedia link structure to generate a Chinese language concept vector, the method comprising: constructing a link information database according to a title concept and/or a link concept in a Chinese-language Wikipedia page; respectively constructing positive training examples and negative training examples with respect to whether a link concept is present in link information database samples, selecting a fixed quantity of positive training examples and negative training examples to create a training dataset; creating a concept vector model, wherein the model comprises an input layer, an embedding layer, a concept vector operation layer and an output layer; and using the training dataset to train the concept vector model and extracting a concept vector from the concept vector model.