Interdisciplinary typically involves the linking or combining of two or more fields of study to create a synthesized whole. Interdisciplinarity is growing rapidly and offers many opportunities, but at the same time the concepts, boundaries and classification of disciplines are not so clear, making it difficult to search and label literature. In recent years, with the development of artificial intelligence technology, the automatic classification of literature has shifted from rule-based classification to machine learning-based classification. This study investigated the automatic classification and recognition of interdisciplinary literatures based on machine learning, taking organic bioelectronics as an example. 8938 literature of organic bioelectronics was collected in Web of Science database from 1980 to 2020 as data source. The approach combining supervised and unsupervised based on machine learning was used to automatically classify and recognize the literatures of organic bioelectronics. A comparative analysis was done for different ML algorithms in terms of accuracy and recall of classification. The results showed over 80% accuracy and recall of literature classification using neural network algorithm. Five topic clusters containing 43 keywords were recognized by unsupervised learning. And the method could be used as one of the effective ways to explore the classification and recognition of interdisciplinary literature.