In ML training and other analytics I often combine features to produce a 'metafeature' and then perform a 'groupby' (pandas) or 'GROUP BY' (SQL) query. What is the technical term for this operation?

Example, in my work I will combine taxonomy, e.g. the genus with a given gene function and gene variant. You could instead consider geographic location of an individual with age and/or occupation, e,g,

location name age first name
California Aderson 34 Jill
NewYork Smith 25 Mark
California Aderson 34 Bob

Group by operation...

Metafeature frequency
California-Aderson-34 2
NewYork-Smith-25 1
  • 1
    $\begingroup$ I just consider this grouping by multiple features. $\endgroup$
    – Ben Reiniger
    Aug 4, 2023 at 20:31
  • $\begingroup$ Thanks @BenReiniger $\endgroup$
    – M__
    Aug 4, 2023 at 21:24


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.