Machine learning systems for automated database element processing and prediction output generation

A computerized method of automatic distributed communication, the method comprising:

• training a first machine learning model with historical feature vector inputs to generate a likelihood output, wherein:
• the historical feature vector inputs include historical profile data structures specific to multiple historical entities,
• the historical feature vector inputs are stored in a Hadoop database or architecture and are passed to a server for use by a Python inference during training, and the likelihood output is populated back to the Hadoop database, and
• the historical profile data structures include structured claim data and structured profile data;
• training a second machine learning model with the historical feature vector inputs to generate a mean count output, wherein training the second machine learning model includes,
• iteratively building the second machine learning model using at least one of a gradient boosting decision tree or regression with a Poisson loss function, to remove variables that are not statistically significant,
• executing early stopping regularization to limit a number of iterations during training of the second machine learning model to prevent overfitting of the second machine learning model,
• applying bagging bootstrap aggregation to repeatedly sample data during training iterations to prevent overfitting of the second machine learning model, and
• implementing five-fold cross validation to split sample data and repeat refitting of the second machine learning model;
• obtaining a set of entities;
• for each entity in the set of entities:
• obtaining structured claim data and structured profile data associated with the entity from a structured profile database;
• generating a likelihood feature vector input according to the obtained structured claim data and structured profile data;
• processing, by the first machine learning model, the likelihood feature vector input to generate the likelihood output, wherein the likelihood output is indicative of a likelihood that the entity will have an avoidable negative health event within a specified first time period;
• selectively including the entity in a first subset of entities based on a comparison of the likelihood output to a likelihood threshold value;
• generating a mean count feature vector input according to the obtained structured claim data and structured profile data;
• processing, by the second machine learning model, the mean count feature vector input to generate the mean count output, wherein the mean count output is indicative of an expected number of avoidable negative health events that the entity will have within a specified second time period;
• and
• selectively including the entity in a second subset of entities based on a comparison of the mean count output to a mean count threshold value;
• and
• automatically distributing structured campaign data to at least one of the first subset of entities and the second subset of entities.