Method and device for detecting document

Abstract

The invention provides a method for detecting a document. The method comprises the following steps of: acquiring paragraph characteristic information corresponding to the document; comparing the paragraph characteristic information of the document with paragraph characteristic information of at least one existing document; and judging whether the existing document similar to the document is provided according to a comparison result. The document can be detected by using the paragraph characteristic information, so the similarities of the documents can be compared accurately, and cheating on sectional processing of the document is avoided; moreover, the checking efficiency is higher and the pressing pressure of a server is lower; the method for detecting the document can be used for improving the online document copyright property detection, and detecting the document when the document is uploaded, so that the following document copyright property detection which causes unnecessary pressure on the server can be avoided; and the copyright property detection of the existing documents can be processed in mass, so the efficiency is higher.

Description

technical field [0001] The present invention relates to a document detection method and device, in particular to a document detection method and device for comparing the similarity of long documents. Background technique [0002] Usually, the document detection method used for document similarity is to confirm through the title, author and word information of the document. However, such an approach has the following disadvantages: First, through the query of the title, author, and word information of the document, omissions are likely to occur, for example, modify or delete the title and author information of the document, or divide the document into multiple parts , so that it is impossible to accurately query or compare other documents through the word information; secondly, if the document to be queried is long, such as a novel, the query is performed through the word information, and the query efficiency is low, and the server or computer processing The pressure is high...

AI Technical Summary

Problems solved by technology

However, such an approach has the following disadvantages: First, through the query of the title, author, and word information of the document, omissions are likely to occur, for example, modify or delete the title and author information of the document, or divide the document into multiple parts , so that it is impossible to accurately query or compare other documents through the word information; secondly, if the document to be queried is long, such as a novel, the query is performed through the word information, and the query efficiency is low, and the server or computer processing The pressure is high, which affects the normal work efficiency of the server or computer

Images