Under the dual background of the construction of the “new liberal arts” and the digital wave, the interdisciplinary practice of combining humanities and technology continues to develop. Taking a number of Chinese language and literature works as examples, this paper selects language features from the vocabulary and sentence levels, analyzes the syntactic structure of the selected Chinese language and literature works with the help of natural language processing technology and numerical measurement method of language features improved TF-IDF method, and realizes the discussion of the lexical categories of literary works, such as word length, word frequency, word class distribution and word density, as well as the study of sentence categories such as average sentence length, sentence dispersion and sentence class distribution. It is found that most of the utterances of the selected literary works are monosyllabic words and polysyllabic words, the cumulative proportion of both of them is more than 90%, the highest frequency of occurrence is nouns and verbs, both of them are more than 22%, the average sentence length and sentence dispersion do not differ much, and the overall readability of the selected literary works is better, with a free change of syntactic structure and a stronger narrative of the text.