Efficient SQL Based Multi-Attribute Clustering

A system, comprising:

• a data storage device configured to store a plurality of data records for a plurality of members, wherein each data record comprises a plurality of data attributes;
• a processor in data communication with the data storage device and configured to:
• calculate, in parallel, a subset outcome value for a plurality of subsets of data attributes of a plurality of data attributes, wherein each subset of data attributes comprises a combination of at least one data attribute of the plurality of data attributes, each subset outcome value representing an average of an outcome for members that are associated with each data attribute of the subset of data attributes, wherein each of the plurality of data attributes are associated with a plurality of members;
• compare, for each subset of data attributes, a number of members associated with the subset of data attributes with a size threshold;
• remove each subset of data attributes for which the number of members associated with the subset is below the size threshold;
• compare, for each subset of data attributes, the subset outcome value for the subset with the global outcome value;
• and
• generate a report that identifies each subset of data attributes for which the corresponding subset outcome value is greater by a first threshold than or less by a second threshold than the global outcome value.

A computer program product, comprising a non-transitory computer readable medium having computer executable instructions to perform operations comprising:

• calculating, in parallel, a subset outcome value for a plurality of subsets of data attributes of a plurality of data attributes, wherein each subset of data attributes comprises a combination of at least one data attribute of the plurality of data attributes, each subset outcome value representing an average of an outcome for members that are associated with each data attribute of the subset of data attributes, wherein each of the plurality of data attributes are associated with a plurality of members;
• comparing, for each subset of data attributes, a number of members associated with the subset of data attributes with a size threshold;
• removing each subset of data attributes for which the number of members associated with the subset is below the size threshold;
• comparing, for each subset of data attributes, the subset outcome value for the subset with the global outcome value;
• and
• generating a report that identifies each subset of data attributes for which the corresponding subset outcome value is greater by a first threshold than or less by a second threshold than the global outcome value.