Systems, methods, computing platforms, and storage media for comparing data sets through decomposing data into a directed acyclic graph

A system configured for decomposing data, the system comprising:

• one or more hardware processors configured by machine-readable instructions to:
• receive an input data set, the input data set including information to be analyzed;
• analyze the input data set to determine a first data subset having a divergent pattern, wherein the first data subset is defined by a first set of filters;
• determine, in a first iteration, one or more first children subsets having the divergent pattern, wherein the one or more first children subsets are each defined by one or more additional filters to the first set of filters;
• determine, in the first iteration, one or more first parent subsets having the divergent pattern, wherein the one or more first parent subsets are each defined by at least one fewer filter than the one or more first children subsets, and wherein the one or more first parent subsets are different than the first data subset;
• determine, in one or more subsequent iterations until a stopping condition is reached, one or more additional children subsets having the divergent pattern, wherein each of the one or more additional children subsets are defined by at least one additional filter to that of previously determined parent subsets;
• determine, in the one or more subsequent iterations until the stopping condition is reached, one or more additional parent subsets having the divergent pattern, wherein the one or more additional parent subsets are defined by at least one fewer filter than previously determined children subsets;
• identify notable subsets within the input data set using one or more heuristic inputs comprising at least one key performance indicator;
• process the identified notable subsets using one or more machine learning techniques to identify subsets or patterns based at least in part on the one or more heuristic inputs;
• and
• recursively repeat identifying and evaluating subsets using the one or more machine learning techniques based on relationships with multiple parent subsets to refine identification of one or more additional notable subsets within the input data set until the stopping condition is reached.