Um Sistema de Prevenção de Vazamento de Dados de Imagens Baseado em Aprendizado de Máquina

Abstract

Technological advancement contributes to the increased risk of loss of sensitive business and household data. Despite the evolution and wide availability of protection tools, in current data leak prevention systems such as Data Leak Prevent (DLP), lack of flexibility, clarity and functional limitations make it difficult to choose. There are several commercial solutions that often have a high cost of licensing, deployment, and limitations of prevention features. In this context, this work proposes and validates the application of a neural network in a DLP proxy to prevent the unauthorized sending of sensitive data stored or captured by video cameras.