Systems and methods for determining network data quality and identifying anomalous network behavior

A device, comprising:

• one or more memories;
• and
• one or more processors, communicatively coupled to the one or more memories, to:
• receive network data associated with a network,
• wherein the network data includes one or more artificial packets,
• wherein the network includes a plurality of network devices, and
• wherein the network data is received from one or more sensors associated with the plurality of network devices;
• process the network data to generate sequential sets of the network data;
• process the sequential sets of the network data, with a first model, to generate time series network data;
• determine whether the time series network data satisfies a data quality threshold;
• receive additional network data when the time series network data fails to satisfy the data quality threshold;
• transform, when the time series network data satisfies the data quality threshold, residual data in the time series network data to identify data outliers in the time series network data;
• remove the data outliers from the time series network data to generate modified time series network data;
• process the modified time series network data, with a second model, to generate a forecast data point;
• compare the forecast data point and an actual data point to determine whether an anomaly exists in the network;
• and
• perform one or more actions based on determining whether the anomaly exists in the network.

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors, cause the one or more processors to:
• receive network data associated with a network,
• wherein the network data includes one or more artificial packets, and
• wherein the network includes a plurality of network devices;
• order the network data to generate sequential network data;
• group the sequential network data into sets to generate sequential sets of the network data;
• process the sequential sets of the network data, with a first time series model, to generate time series network data,
• wherein the first time series model is trained with historical network data prior to receiving the network data;
• determine whether the time series network data satisfies a data quality threshold;
• transform, when the time series network data satisfies the data quality threshold, residual data in the time series network data to identify data outliers in the time series network data;
• remove the data outliers from the time series network data to generate modified time series network data;
• process the modified time series network data, with a second time series model, to generate a forecast data point,
• wherein the second time series model is trained based on the historical network data prior to receiving the network data;
• compare the forecast data point and an actual data point to determine whether an anomaly exists in the network;
• and
• perform one or more actions based on determining whether the anomaly exists in the network.