Analysis of digital humanistic knowledge production based on natural language processing

Abstract Digital humanistic knowledge production emphasises the importance of a strong community and differentiates from traditional models, which include aspects such as online cooperative development. The digital humanities model is already widely acknowledged. However, features characteristics under natural language processing are controversial. This research presents wordVEA feature mining approach based on word2vec variational self-encoder (VAE). primarily defined by coexistence structure boundary blurring, well interdisciplinary collaboration thematic cohesiveness broad horizon, determined results effectively address question through application word VAE method.