English literature occupies an important position in the world literary world and shows a high literary research value. The study explores English literature from a cross-cultural perspective and systematically analyses the linguistic features of English literature. Then, it introduces the digital humanities method, combines the Word2vec algorithm and TF-IDF algorithm for text representation and feature extraction, and constructs a model for analyzing linguistic features. Taking the English literary works of two certain writers as an example, text features and text emotions are interpreted to uncover the linguistic expression features of English literature. English literature as a whole reflects the features of quoting classics, reality-based, cultural differences, dramatic monologues, and European cultural characteristics. The readability of the sample English literature was at the slightly difficult and difficult levels, with scores of 45.53 and 52.85, respectively, and Author A’s English literature was of higher complexity and had a higher degree of textual subordination (1.35). The selected English literature samples had mainly neutral (47.87%) and positive (40.61%) affective tendencies. With the ever-accelerating cultural globalization and exchange, there is a proliferation of new and old outstanding works from all ethnic regions of the world, which can only be better interpreted by fully respecting cultural differences and improving literary literacy.