(pour un public de progammeurs)
Double intention
feature engineering is another topic which doesn’t seem to merit any review papers or books, or even chapters in books, but it is absolutely vital to ML success. […] Much of the success of machine learning is actually success in engineering features that a learner can understand.
Processus consistant à utiliser la connaissance du domaine pour extraire des features en transformant une partie des données brutes par des méthodes de data mining
Améliorer les performances de vos algorithmes
Rendre le data set plus facilement interprétable (machine et humain)
Scott Locklin, in “Neglected machine learning ideas”
méthodes non supervisées
Label propagation
PCA
SVD
voir blog kaggle