Traffic monitoring is an important task for network administrators who require tools to aid in the detection of changes in the network’s routine. In this paper, we use a Digital Signature of Network Segment using Flow Analysis (DSNSF) as a technique to describe standard network behavior aiming to support network management through traffic characterization. We have collected real data set from State University of Londrina (UEL), using data flow attributes such as bits, packets and number of flows. Our novel model uses Genetic Algorithm to optimize the process, which consists or organizing the data to display graphically a standard network behavior. To accomplish this task, we compared our novel model with another similar method, Ant Colony Optimization for Digital Signature (ACODS), evaluating these models to measure their accuracy.