Abstract: In order to manage tobacco literature knowledge high efficiently and accelerate the transformation of tobacco literature resources from data security type to knowledge service type, a retrieval service platform for tobacco literatures based on B/S architecture was configured by using big data analysis and natural language processing technologies and following J2EE development standard specifications. Through combining advanced retrieval technology and knowledge discovery algorithm with knowledge graph visual analysis technology, the platform could deeply integrate and mine tobacco literature resources and realize intelligent retrieval, intelligent recommendation, intelligent analysis, scientific evaluation and other functions. The service platform improves the accuracy of tobacco literature resource acquisition and promotes the transformation of tobacco scientific and technological achievements effectively.