Feature Engineering
هندسة الميزات
Definition
Designing, transforming and selecting a model's input variables from raw data to improve its performance.
Definition
Feature engineering is the process of using domain knowledge to create, transform and select input variables from raw data so that machine learning models perform better.
Practical examples
import pandas as pd
df = pd.DataFrame({"city": ["Baghdad", "Basra", "Baghdad"], "age": [20, 30, 40]})
df = pd.get_dummies(df, columns=["city"], dtype=int)
df["age_z"] = (df["age"] - df["age"].mean()) / df["age"].std(ddof=0)
print(df.round(3).to_string(index=False))Expected output
age city_Baghdad city_Basra age_z 20 1 0 -1.225 30 0 1 0.000 40 1 0 1.225
Mathematical equations
z = \frac{x - \mu}{\sigma}Scientific sources
References and scientific sources
-
ArticleA Few Useful Things to Know About Machine Learning An article on overfitting, feature engineering and model generalization.
-
BookThe Elements of Statistical Learning (2nd ed.) A statistical reference on machine learning, freely available from the authors' website.
