Document vectors should be compared by direction rather than total length. Why use Normalizer instead of StandardScaler?
Classic MLLevel: IntermediatePer-sample normalization
Document vectors should be compared by direction rather than total length. Why use Normalizer instead of StandardScaler?
Hints
- Decide whether rows or columns are normalized.
- Direction survives positive row rescaling.
Answer
Normalizer rescales each sample to unit norm. StandardScaler rescales columns using population statistics; it does not enforce equal document length.
Source and licence: scikit-learn 1.7.2 — preprocessing · BSD-3-Clause · answer: Quantoros editorial