Complex text recommendation method and system based on semantic enhancement

The invention discloses a complex text recommendation method and system based on semantic enhancement, and aims to solve the problem of complex achievement description text supplied in an achievement transformation platform, and provides a method for carrying out content extension on text feature words by adopting a general vocabulary text extension mode on the basis of complex text feature extraction and aiming at the problem of Chinese vocabulary diversity. And a context feature extraction algorithm is used to fuse structure information of resource entities and extended text information, association information between entities is established based on historical interaction information, recommendation model training is performed through a recommendation algorithm based on fused entity vectors, and intelligent recommendation in a complex text description scene is realized. The recommendation model is constructed for the recommendation of the complex text, and the efficiency and accuracy of the recommendation