Multi-objective recommendations in a data analytics system

A non-transitory machine-readable storage medium that provides instructions that, if executed by a processor of a computing device, will cause the computing device to perform operations for providing multi-objective recommendations, the operations comprising:

• storing a data set including a plurality of records, wherein each of the plurality of records includes a plurality of fields;
• receiving user input indicating a plurality of objectives, wherein each of the plurality of objectives indicates a desired goal for a field of interest;
• receiving user input indicating a plurality of actionable fields;
• receiving user input indicating selection of one of the plurality of records in the data set;
• determining one or more candidate changes to values of the plurality of actionable fields of the selected record, wherein the one or more candidate changes are determined based on applying an evolutionary algorithm to a population of previously observed values of the plurality of actionable fields to find a non-dominated set with respect to the plurality of objectives;
• determining, for each of the one or more candidate changes, a multi-objective score for that candidate change based on predicted values of the fields of interest indicated by the plurality of objectives if that candidate change is made;
• selecting one or more of the one or more candidate changes to recommend to a user based on the multi-objective scores of the one or more candidate changes;
• and
• providing, for display to the user, the selected one or more candidate changes as recommended changes.

A computing device configured to provide multi-objective recommendations, the computing device comprising:

• one or more processors;
• and
• a non-transitory machine-readable storage medium having instructions stored therein, which when executed by the one or more processors, causes the computing device to:
• store a data set including a plurality of records, wherein each of the plurality of records includes a plurality of fields,
• receive user input indicating a plurality of objectives, wherein each of the plurality of objectives indicates a desired goal for a field of interest,
• receive user input indicating a plurality of actionable fields,
• receive user input indicating selection of one of the plurality of records in the data set,
• determine one or more candidate changes to values of the plurality of actionable fields of the selected record, wherein the one or more candidate changes are determined based on applying an evolutionary algorithm to a population of previously observed values of the plurality of actionable fields to find a non-dominated set with respect to the plurality of objectives,
• determine, for each of the one or more candidate changes, a multi-objective score for that candidate change based on predicted values of the fields of interest indicated by the plurality of objectives if that candidate change is made,
• select one or more of the one or more candidate changes to recommend to a user based on the multi-objective scores of the one or more candidate changes, and
• provide, for display to the user, the selected one or more candidate changes as recommended changes.