特征工程是机器学习中最重要的环节。好的特征比好的模型更重要。 ## 数值特征处理 ```python from sklearn.preprocessing import StandardScaler, MinMaxScaler scaler = StandardScaler() df['feature_scaled'] = scaler.fit_transform(df[['feature']]) df['feature_log'] = np.log1p(df['feature']) df['age_bin'] = pd.cut(df['age'], bins=[0, 18, 35, 50, 100]) ``` ## 类别特征处理 ```python df = pd.get_dummies(df, columns=['city'], drop_first=True) target_mean = df.groupby('city')['target'].mean() df['city_target_enc'] = df['city'].map(target_mean) ``` ## 特征选择 ```python from sklearn.ensemble import RandomForestClassifier rf = RandomForestClassifier(n_estimators=100) rf.fit(X, y) importance = pd.Series(rf.feature_importances_, index=X.columns) ``` 特征工程是数据科学家的核心竞争力。