Feature Engineering Techniques (Imputation (Types (Numerical (ex:Column…
Feature Engineering Techniques
ex:Column with values '1' & N/A.
Possible that N/A's are supposed to be '0'
Replace missing value with "most occurred"
or "Other" if uniformly distributed
Assigning relevant values missing data
Data point that differs significantly from other observations (More than 3 times the standard
Drop or Cap ?
Transform numerical values to categorical eqivalents
Transform highly skewed data to less skewed
Categorical to numeric
Grouping repeated multiple
instances(row) in to one row
Key Factor: Identify the right aggregation
Split data to be more meaningful for models
e.g: Full name to "First", "Middle" & "Last"
To makeup for missing range in the data
Important for algorithms based on distance of date (K-NN, K-Means)