問題文
Provider identifier takes about eleven thousand distinct values, and a few hundred of them account for most claims. Expanding it into one column per value would add eleven thousand columns. Which combination is correct here?
選択肢
- Expanding every value would cause the cardinality problem, so the practical route replaces the identifier with the grouped summary that fitting documentation must compute inside each fold.
- Expanding every value would cause the cardinality problem, so the practical route replaces the identifier with the positional summary that fitting documentation must compute inside each fold.
- Expanding every value would cause the precision problem, so the practical route replaces the identifier with the positional summary that fitting documentation must compute inside each fold.
- Expanding every value would cause the precision problem, so the practical route replaces the identifier with the grouped summary that fitting documentation must compute inside each fold.