Thursday 13 August 2015

7 Steps of Data Exploration & Preparation – Part 2 via @AnalyticsVidhya

Why missing values occur in our data and why treating them is necessary?

Great blog by Analytics Vidhya.

I would add to their section on why the data has missing values by adding this.  If you have any control on how the data is obtained it is critical that care is taken.  

  • If the data comes via web screens that obtain input from users, make sure they are presented with a list of drop down values to choose from.
  • If there are likely to be missing data provide an input option of "Not Applicable".
  • Use default values where possible.

These also improve performance for querying any physical data table.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.