Please enable JavaScript.
Coggle requires JavaScript to display documents.
Creating New Data Columns (Text Data (Splitting and Extracting New…
Creating New Data Columns
Text Data
Splitting and Extracting New Columns
IF-THEN Statement : Examining a value in a column and making changes to this value and other values elsewhere
One-Hot Encoding:Encoding integer feature of values
Categorical Data
Numerical Data
Transforming
Natural Logarithm ( log() ) : Linearizing exponential data
Exponentiation ( ** ) : Might used when calulating financial data
Equal ( = = ) : Comparing when two data points are the same and might cancel out each other or may indicate a higher likelihood of a target phenomenon
Not Equal ( !=) : Comparing when two data points are not the same
Greater Than or Equal ( >= )
Greater Than ( > )
Less Than or Equal ( <= )
Less Than ( < )
Divison ( / ) : Dividing one column by another, and might make hidden information from some types of algorithms available
Multiplication (*) : Used when two related columns interact with a target in a way that is only detectable through their product
Absolute ( Abs() ) : Used to calculate the distance of a number on the number line from 0, without negative number
Subtraction (-) : Subtracting one column from one another to increase the similarity or difference between values
Addition (+) : Adding different columns to increase predictive signals
Square Foot ( Sqrt() ) : Works for different distribution of data
Square ( Square() ) : Makes large values more larger