Precise Information Extraction from Text Based on Two-Level Concept Lattice

Abstract

Aiming at the problems of high cost of time consumption and low precision of IE (information extraction) in IE systems based on concept lattice, a novel mechanism of two-level concept lattice-based IE is put forward. The structure concept lattice is the logical storage of semantic structure of documents, and the content concept lattice is used to store content information of documents. The paper gives the formal descriptions of structure and content concept lattice and analyses the time complexity of them. The consequence shows our method has obvious advantages concerning the time consumption and the precision of content extraction compared with existing IE methods.