Processor and method of weighted feature importance estimation

An apparatus comprising:

• a memory;
• and
• a processor coupled to the memory and configured to:
• determine a feature importance ranking for each pair of clusters of a plurality of clusters associated with data to generate a first plurality of feature importance rankings;
• determine a feature importance ranking between a particular data element of the data and each cluster of the plurality of clusters to generate a second plurality of feature importance rankings;
• determine a distance value associated with each pair of clusters of the plurality of clusters to generate a plurality of distance values, wherein a particular distance value associated with a particular pair of clusters of the plurality of clusters indicates a distance between a first cluster of the particular pair of clusters and a second cluster of the particular pair of clusters;
• for each data element of the data that is within one or more clusters of the plurality of clusters, determine a probability value associated with the data element to generate a plurality of probability values;
• weight the first plurality of feature importance rankings based on the plurality of distance values to determine a first plurality of weighted feature importance rankings;
• and
• weight the second plurality of feature importance rankings based on the plurality of probability values to determine a second plurality of weighted feature importance rankings.

A non-transitory computer-readable medium storing instructions executable by a processor to perform operations, the operations comprising:

• determining a feature importance ranking for each pair of clusters of a plurality of clusters associated with data to generate a first plurality of feature importance rankings;
• determining a feature importance ranking between a particular data element of the data and each cluster of the plurality of clusters to generate a second plurality of feature importance rankings;
• determining a distance value associated with each pair of clusters of the plurality of clusters to generate a plurality of distance values, wherein a particular distance value associated with a particular pair of clusters of the plurality of clusters indicates a distance between a first cluster of the particular pair of clusters and a second cluster of the particular pair of clusters;
• for each data element of the data that is within one or more clusters of the plurality of clusters, determining a probability value associated with the data element to generate a plurality of probability values;
• weighting the first plurality of feature importance rankings based on the plurality of distance values to determine a first plurality of weighted feature importance rankings;
• and
• weighting the second plurality of feature importance rankings based on the plurality of probability values to determine a second plurality of weighted feature importance rankings.