Time Series Feature Engineering

Fire Insights provides a number of Processors for Feature Engineering of Time Series Data. These include:

Update New features where needed
Features Description
DateTimeFieldExtract Extracts year, month, day of month, hour, minute, second and week of year from timestamp/date columns
Days to holiday Days remaining for next holiday
Days from holiday Days passed after holiday
Time-segmentation Divide data in morning, afternoon, evening, night to get more idea about time based pattern
MovingWindowingFunctions Calculates the moving values using the given function
WindowingAnalytics Implements window functions is mainly through the operators rolling and expanding
Exponential Moving Average (EMA) The Exponential Moving Average (EMA) assigns a greater weight to the most recent price observations. While it assigns lesser weight to past data, it is based on a recursive formula that includes in its calculation all the past data in our price series.

DateTimeFieldExtract

Below is the sample workflows which contains DateTimeFieldExtract processor in Fire Insights.

It reads the JetRail Train dataset & use DateTimeFieldExtract processor which create New DataFrame by extracting Date & Time field and print the result.

ml_userguide

DateTimeFieldExtract processor Configuration:

ml_userguide

Output result of DateTimeFieldExtract processor:

ml_userguide

MovingWindowingFunctions

Below is the sample workflows which contains MovingWindowingFunctions processor in Fire Insights.

It reads the ticker dataset, concatenate the input column, casting specified column to new data type, use MovingWindowingFunctions processor which calculates the moving value of selected function of input column and print the result.

ml_userguide

MovingWindowingFunctions processor Configuration:

ml_userguide

Output result of MovingWindowingFunctions processor:

ml_userguide