Abstract: 【Purpose/significance】As a cross-border discipline,Digital Humanities is developing rapidly while facing the problems of
unclear concept definition and lack of special journals,which increase the difficulty of literature collection. It is necessary to build a
suitable identification and classification model, form a Digital Humanities literature library, and help Digital Humanities research.
【Method/process】Analyze the essential of Digital Humanities,summarize its characteristics,and construct machine learning model on
the basis of manual reading annotation to identify and classify Digital Humanities literature automatically.【Result/conclusion】Raise
a classification model of Digital Humanities literature based on machine learning,which achieves good identification effect for Digital
Humanities literature in the field of Library and Information Science.【Innovation/limitation】Apply machine learning algorithm to Digi?
tal Humanities document classification,which can better deal with the problems of complex vocabulary and small amount of data; Fur?
ther research can use more complex models such as deep learning,and realize the multidisciplinary classification of Digital Humanities
documents in different fields.