Ingestion and segmentation of real-time event data

A computing device comprising:

• one or more non-transitory memories;
• and
• one or more processors in communication with the one or more non-transitory memories and configured to execute instructions stored in the one or more non-transitory memories to:
• access, by a management engine, a plurality of subsets of instances, wherein each subset of the plurality of subsets of instances corresponds to a time period;
• identify, by the management engine, a net capacity change for each subset of instances of the plurality of subsets of instances, wherein the net capacity change of a respective subset of instances is based at least in part on a capacity change of each instance in the respective subset of instances;
• generate, by a modeling engine, feature vectors for the plurality of subsets of instances, wherein a respective feature vector comprises information identifying the respective subset of instances;
• and
• train, by the modeling engine and using the feature vectors, a machine learning model to output a predicted net capacity change for a particular entity at an input time period.

A non-transitory computer-readable medium storing a plurality of instructions that, when executed by one or more processors of a computing device, cause the one or more processors to:

• access, by a management engine, a plurality of subsets of instances, wherein each subset of the plurality of subsets of instances corresponds to a time period;
• identify, by the management engine, a net capacity change for each subset of instances of the plurality of subsets of instances, wherein the net capacity change of a respective subset of instances is based at least in part on a capacity change of each instance in the respective subset of instances;
• generate, by a modeling engine, feature vectors for the plurality of subsets of instances, wherein a respective feature vector comprises information identifying the respective subset of instances;
• and
• train, by the modeling engine and using the feature vectors, a machine learning model to output a predicted net capacity change for a particular entity at an input time period.