Construction method of information auditing classification model and information auditing method

Abstract

The invention relates to the technical field of artificial intelligence, and provides a construction method and device of an information auditing classification model, computer equipment and a storage medium. The method comprises the steps of acquiring a sample image, performing optical character recognition on the sample image to obtain a text recognition result, performing semantic analysis, extracting a target text according to a semantic analysis result of the text recognition result, and generating an auditing label corresponding to the target text based on an auditing label generation rule corresponding to a business party to perform auditing label labeling, and performing model training according to the target text carrying the auditing label to obtain an information auditing classification model. Labeling processing is carried out without depending on manpower, the generation speed of samples is increased, audit labels suitable for model training of different business parties can be generated, reuse of sample images is achieved, and an information audit classification model is obtained through rapid training. In addition, the invention also relates to a blockchain technology, and the information auditing result of the user can be stored in the blockchain.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a method for constructing an information review classification model, an information review method, a device, a computer device, and a storage medium. Background technique [0002] With the development of information technology, information review is more and more widely used in daily life. Taking loan review as an example, the lender's application information is reviewed layer by layer, and the bank then conducts manual review. The manual review process is complex and time-consuming, and it is difficult to control the amount of review to be completed on time, causing lenders to question the bank's services. damage the interests of the bank. Due to the large number of processes and personnel involved in information review, the requester waits too long for the information review process. For the reviewer, the review process is very complicated, and it i...

AI Technical Summary

Problems solved by technology

For the reviewer, the review process is very complicated, and it is difficult to judge whether the manual review meets the standard

[0003] With the development of artificial intelligence technology, training neural network models to replace manual processing has been applied in various industries, but in terms of information review, since the training of the model depends on a large number of training samples, and the samples need to be manually processed based on historical review records The sample label is marked, and different business parties have different audit standards, and the model obtained by the same sample training cannot be applied to different business parties. As a result, model training requires a lot of time for sample preparation, and the model training process is inefficient.

Images