Comparative Study of Digital Transformation Based on Neural Network Word Embeddings

• 1、School of Management, Wuhan University of Science and Technology, Wuhan, 430065, China

Abstract：With the advent of the era of digital intelligence, digital transformation has gradually become the only way for enterprises to further develop, and it also plays a vital role in the high-quality development of our economy, which has attracted widespread attention and research in the society. This paper compares the two corpora of digital transformation of the Chinese academic platform CNKI and the social question-answering platform Zhihu, aiming to promote the research of digital transformation in China to achieve more results. Firstly, the emerging neural network word embedding algorithm is used to model the word turn value Word2vec of the two corpora, so as to measure the similarity of words effectively at both the grammatical and semantic levels. Then, the top 20 words that are most similar to "digital transformation", the top 20 high-frequency words and the top 10 words that are most similar to "enterprise" are obtained for comparative analysis, and t-SNE reduction and matplotlib visualization methods are used to compare and analyze the word embedding overview of the words. The research results show the differences between Chinese academic platform and social question-and-answer platform in digital transformation research. This paper innovatively uses Word2vec to compare and analyze the research of digital transformation between Chinese academic platform and social question-and-answer platform, which provides a new perspective for the research of digital transformation.

Keywords： digital transformation Word2vec academic platform social question-and-answer platform comparative study