Analyze the subordination structure between domain-specific vocabulary and meaning with the Word2Vec training process

Word2vec uses the relationship between the occurrence of words before and after a sentence as the basis for training, which is similar to the human habit of using words. A single word2vec model has only a fixed word vector space and can only rely on a single word vector similarity for clustering, so the word2vec training process will begin to gradually disperse the word vector space and tend to homogeneous phenomenon for clustering assistance. During training, words that are more similar are dispersed more slowly than other words because the possibility of subgroups with a closer semantic relationship can be observed. In addition, the data used in the experiments of this study are Chinese datasets.