基于连续空间表示的文本分类

he representation of text contiguous space is representing the characteristics of the text by a distributed vector.the experiment of this paper represent a text by a vector through the training of doc2vec model.And the vector got from this model have syntax information and with semantic information.What's more,the vectors of the texts with similar semantic distance as closely as possible.Because text classification is a benchmark of checking a algorithm in the application of natural language processing,so the purpose of this paper is training text vector first,then applying it in text classification,and determining the training effect according to the result of text classification.This experiment aim to apply text vector to text classification, and get a good result.