Sensitive data decision judging method and system

Abstract

The invention discloses a sensitive data decision judging method and system. The system comprises a decision judging device, a document input device, a judging result confirmation device and a feedback device, wherein the decision judging device comprises a plurality of sensitive data judging methods and is used for judging whether a to-be-judged document comprises sensitive data or not and determining a weight, in the decision judging, of each sensitive data judging method according to history judging conditions of each sensitive data judging method; the document input device is used for inputting the to-be-judged document to the decision judging device; the judging result confirmation device is used for confirming correctness of a judging result formed by the decision judging system, wherein the confirmed result comprises mistaken report and normal; and the feedback device is used for feeding the mistaken report result back to the decision judging device, re-determining a weight, inthe decision judging, of each sensitive data judging method, and updating the decision judging system. Through the method and system provided by the invention, sensitive data recognition result is more correct, the mistaken report rate and missing report rate are decreased, and various types of documents can be effectively recognized and judged.

Description

technical field [0001] The invention relates to the field of data security, in particular to a sensitive data decision-making method and system. Background technique [0002] In recent years, due to the increasingly diversified formats and storage forms of electronic data within enterprises, the risk of enterprise sensitive data leakage has increased sharply, and enterprises have paid more and more attention to the prevention of their own sensitive data leakage. However, the formats of enterprise documents are diverse, and how to discover and identify sensitive data has become a major problem faced by enterprises. The sensitive data of an enterprise is mainly reflected in the data content, so content-based sensitive data identification technology has become the key technology of DLP, the mainstream data leakage prevention product in the market. Content-based sensitive data discovery technology mainly includes [0003] There are three parts: document format recognition, doc...

AI Technical Summary

Problems solved by technology

[0011] (1) In the process based on a single identification technology, the false negative rate and false negative rate show a seesaw effect, and it is actually difficult to reduce the false negative rate or false positive rate in one direction

[0012] (2) The maximum value principle cannot comprehensively correlate the conclusions of multiple detection methods

[0013] (3) Different recognition methods have their own advantages and disadvantages in different document types and policy contexts, and the simple maximum principle cannot integrate the advantages and disadvantages of each method

[0015] First of all, the regular keyword recognition method, the file fingerprint recognition method, and the VSM vector space model recognition method have high recognition accuracy for document types that are suitable for the method itself. It is difficult to control the false positive rate when the keyword regular recognition method detects text documents

Images