Open type document isomorphism engines system

Abstract

An open document isomorphism engine system in the field of information security technology, wherein: the physical structure module accepts the input of various documents, and outputs the physical structure of the document to the logical structure module; the logical structure module processes the information input by the physical structure module Process to obtain the logical structure of the document, and input it to the lexical and syntactic analysis module; the lexical and syntactic analysis module receives the information input by the logical structure module, and processes the information to obtain the analyzed document, and the obtained The document is input into the concept extraction module; the concept extraction module processes the information input by the lexical and syntactic analysis module to obtain the concept and concept attributes converted from the words in the document, and inputs the obtained concepts and concept attributes into the topic representation module; topic The representation module processes the information input by the concept extraction module to obtain the document theme in concept units. The invention solves the problem that multi-format documents cannot be processed uniformly.

Description

technical field [0001] The present invention relates to a system in the technical field of information security, in particular to an open document isomorphic engine system (ODIE-Open Document Isomorphic Engine). Background technique [0002] In the field of content security, all content security products based on text information must perform semantic understanding of text and filter bad information. Such products all face a unified problem, that is, to extract plain text information for understanding and filtering from various documents. Due to the complexity and diversity of document formats in reality, most systems avoid this difficult problem, resulting in low accuracy of these systems. [0003] At present, there are two difficult problems in the process of obtaining plain text information: (1), how to deal with various original document formats and obtain plain text information therefrom. According to the degree of structure, various electronic documents in reality ca...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the deficiencies of the prior art, to provide an open document isomorphism engine system, which can be used to extract the plain text content of multi-format documents and the semantics it represents, and solve the problem that cannot be solved for multi-format documents. Unified processing of issues that can be applied to semantic and Internet content security analysis projects

Images