Document auditing method, device and system, equipment and storage medium

Abstract

The embodiment of the invention discloses a document auditing method, device and system, equipment and a storage medium. The method comprises the following steps: acquiring an object numbered musicalnotation format document sent by a document analysis component, and acquiring a preset machine learning model and a document auditing rule matched with the object numbered musical notation format document according to a business scene of the object numbered musical notation format document; extracting key information in the object numbered musical notation format document through a preset machinelearning model; and generating a document auditing result through a rule engine according to the document auditing rule and the key information, and sending the document auditing result to the webpagecomponent. According to the embodiment of the invention, the matched preset machine learning model and document auditing rule can be obtained according to the business scene. The key information in the object numbered musical notation format document can be accurately extracted through the preset machine learning model, and then document auditing is carried out through the rule engine according to the document auditing rule and the key information, so that the document auditing requirements under different business scenes are met, and the document auditing effect and the document auditing efficiency are improved.

Description

technical field [0001] Embodiments of the present invention relate to document processing technology, and in particular to a document review method, device, system, equipment and storage medium. Background technique [0002] Document review is a common work scenario. Various types of documents have related review standards and procedures, and only compliant documents can enter the next step. At present, most of the document review work is purely manual. Not only is the efficiency relatively low, but also due to the boring and heavy work, some information may be missed, leading to risks. [0003] The existing technology introduces automation solutions in some links of document review, which can perform certain auxiliary work in a certain link, reducing labor costs and pressure. For example, the document review process includes: document parsing, information extraction, and document review. The document review system in the prior art extracts information based on traditiona...

AI Technical Summary

Problems solved by technology

[0004] The defect of the existing technology is that information extraction based on traditional technologies such as keyword search, text matching, and regular expressions, often due to reasons such as rich text expressions in actual business forecast samples, leads to poor information extraction functions and directly affects subsequent links.

At the same time, due to the particularity of the field and the difference of document categories, the document review system in the prior art often can only deal with a certain scene of a certain document in a certain field

Between different scenarios, the document review system is quite different and cannot be expanded

New scenarios often need to be developed from scratch, the existing system is less reused, and the development cycle is longer

Images