Because the detection of malicious files is not yet popular, there are often various "backdoors" for files on most forums. For Internet users who publish other people's privacy, bad comments or pictures in cyberspace, and upload and download malicious files in various forums and communities, this paper proposes that through the use of naive Bayesian algorithm and AC computer algorithm, we can quickly and accurately identify malicious content in text, pictures, and software. Through the secondary development of API, we can monitor hundreds of millions of data in real-time, It greatly alleviates the problem of detecting malicious files in the current environment.
|