Computing processes with data and file management control systems for database classification

A computing system for process management and control for database classification, comprising:

• a computer with one or more processor and memory that stores computer-readable instructions;
• and
• wherein, upon execution, by the one or more processor, of the computer-readable instructions, the computer performs steps comprising:
• accessing a data frame;
• applying predetermined rules for reconfiguring data in the data frame to a standard format, the predetermined rules including at least replacing null values with predetermined default values;
• training a plurality of machine learning models on the data to generate a plurality of candidate rules, wherein the training includes:
• inserting training data into an iterative training and testing loop to predict a target variable;
• using features in the training test data to train the model based on weights and iterative calculations;
• and
• repeatedly predicting the target variable during each iteration of the training and testing loop, wherein each iteration of the training and testing loop has differing weights, each of the differing weights being updated with each iteration of the training and testing loop to reduce error in predicting the target variable, until the model surpasses a predetermined minimum threshold level of accuracy, wherein updating the weights includes:
• comparing a generated output produced by a neural network in response to the training data with a desired output;
• determining an associated error amount for each of a plurality of output data points generated in an output layer;
• communicating the associated error amount back through the system as an error signal having a value between −1 and 1, and
• adjusting the weights based on the error signal;
• computing each of the candidate rules using the data as input, and generating a score for each of the candidate rules;
• generating an output file including the candidate rules and the associated scores;
• storing the output file for further use;
• selecting at least one of the candidate rules;
• and
• implementing the selected at least one candidate rule in a software application of an enterprise computing system.