Document comparison method and device, computer equipment and storage medium

Abstract

The invention relates to a document comparison method and device, computer equipment and a storage medium. The method comprises the following steps: calculating the credibility of characters in a target document, and replacing the characters of which the credibility is lower than a preset threshold value with predicted characters conforming to a screening condition to obtain an error correction document; identifying the content in the error correction document, and determining the format type of the content; performing semantic distance calculation on the content of the format category in the error correction document and the content corresponding to the format category in a reference document, and determining a minimum unit needing to be compared in the content of the format category; and comparing the content in the minimum unit with the content corresponding to the minimum unit in a reference document to obtain a comparison result. By adopting the method, the processing of the editable electronic document and the image scanning file can be realized, and the anti-counterfeiting comparison is carried out on the contents of the editable electronic document and the image scanning file, so that the duration of the anti-counterfeiting comparison and auditing is shortened, and the risk of auditing omission is reduced.

Description

technical field [0001] The present disclosure relates to the technical field of document anti-counterfeiting, in particular to a document comparison method, device, computer equipment and storage medium. Background technique [0002] In the credit business scenario of the financial industry, business personnel need to process a large number of loan contracts every day. The contract generally uses a standard contract. In order to prevent the contract from being tampered with by the other party and causing disputes, the issuer of the contract needs to review and confirm all the textual terms of the contract. Due to the constraints of the business process, in the traditional business working mode, auditors need to proofread the scanned images one by one to ensure that the contract content is complete and correct. [0003] The current main method of contract comparison is based on the experience of industry or business experts. The points that need to be compared for the same c...

AI Technical Summary

Problems solved by technology

[0003] The current main method of contract comparison is based on the experience of industry or business experts. The points that need to be compared for the same contract are filtered by configuring the rule engine, so as to realize the comparison of two contracts. The main The disadvantage is that it is necessary to exhaustive rules for different types of contracts in order to complete the ability to compare all contracts

Another way of contract comparison is to use the sequence labeling method, combined with the understanding of data and business characteristics, to build a comparison model, so as to realize the comparison of two contracts, but the effect of this method is poor, the current business and Strong data personalization, it is difficult to complete the collection of massive data

Images