updated

Author: peicheng

FeatureEngineering.md

@@ -1,8 +1,13 @@
 # Feature Engineering
-Yin-Yang sides of feature engineering
+Yin-Yang sides of feature engineering.
+
 
 ## Transforming or creating new features (Yang-side)
 
+We can combine tables, transform features to create new features.
+We can even do so with help. Let's see [featuretools](https://www.featuretools.com/)
+
+After we created "meaningful" features, we are now ready to transform their format.
 ### OneHot Encoding: Dealing with discrete features
 
 ### Mathematical Operations: Creating New Features
@@ -17,10 +22,15 @@ Apply product,ratio, or other transformations that take 2 or more inputs.
 
 ### NLP: dealing with text feature
 
-## Feature selection (Yin-side)
+## Dimension Reduction (Yin-side)
+
+### Principal Component Analysis (PCA)
+
+### auto-encoding using deep neural networks
 
-### Selection before traning
+### Feature selection
 
-### Selection while training - LASSO regularization
+Before the final task, we could try to solve a representive task. Use feature importance for a model, usually trees, to select features. Use SelectKBest (e.g. [here](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectKBest.html)).
+In addition, we can do selection while training - LASSO regularization etc.
 
 

README.md

@@ -35,7 +35,7 @@ There are many great sources to learn Data Science, and here are some advice to
         * Take a First Look
         * Data Cleaning
         * Transform DataFrames
-    * Dimension Reduction
+    * [Feature Engineering](FeatureEngineering.md)
 
 4. [Exploring Data](ExploringData.md)
     * Simple Data Visualization
@@ -50,7 +50,6 @@ There are many great sources to learn Data Science, and here are some advice to
     * Hierarchical clustering
 
 7. [Basic Supervised Learning](SupervisedLearningBasic.md)
-    * [Feature Engineering](FeatureEngineering.md)
     * [Classification via Logistic Regression](https://www.kaggle.com/danielzou/tensorflow-multiclassification)
     * Ensemble Learning
         * [XGBoost](XGBoost.md)