Risk prompting method, information prompting method, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a risk prompting method. The method comprises the following steps of obtaining text elements in the target text, determining first risk data of the text elements, determining that the target text has a risk, according to the first risk data, the risk prompt being carried out on the at least one target text element so that the at least one target text element which enables the target text to have the risk can be prompted. The method is advantaged in that the part with the risk can be positioned more accurately, a problem of missing report or false report when the target text element is searched manually is avoided, and working efficiency and quality are improved.

Description

technical field [0001] The present application relates to the technical field of data processing, and in particular to a risk notification method, an information notification method, a computer device, and a computer-readable storage medium. Background technique [0002] With the rapid development of the Internet and the rapid increase in the amount of information, a large amount of pornographic, violent, illegal and other harmful information is mixed in it, seriously affecting the healthy development of the Internet. For Internet platforms, a large number of users are constantly producing content every day, which requires a very high cost for content review. [0003] At present, content review needs to rely on a lot of manpower to review the content generated by the website one by one, and make appraisal conclusions based on the content. However, the rapid growth of Internet information makes it difficult for manual review to balance both accuracy and work efficiency. Art...

AI Technical Summary

Problems solved by technology

However, the rapid growth of Internet information makes it difficult for manual review to balance both accuracy and work efficiency

Artificial intelligence content auditing emerged as the times require, but it can only help manual rough screening, screening out risky articles, pictures or videos, etc., reducing part of the manual workload

[0004] The applicant has found through research that artificial intelligence content review is usually a black box, which can only screen out risky articles, but there are still many articles that are screened out, which need to be reviewed manually, and there are still a large number of false negatives and false positives. problems, the efficiency and quality of audits are difficult to guarantee

Images