Predictive Analytics Times
Predictive Analytics Times
EXCLUSIVE HIGHLIGHTS
Dr. Data Show Video: How Can You Trust AI?
 Watch the second episode of The...
Three Common Mistakes That Can Derail Your Team’s Predictive Analytics Efforts
  Originally published by Harvard Business...
Artificial Intelligence: Are We Effectively Assessing Its Business Value?
 As most data science practitioners know,...
Dr. Data Show Video: Why Machine Learning Is the Coolest Science
 Watch the premiere episode of The...
SHARE THIS:

2 years ago
Ghosts in the Data, Constructing Data Entities

 

Data Entities are seldom discussed concepts that primarily hide in the shadows or are ghosts on the periphery. These entities are data constructs that are observationally defined in terms of the underlying data set that can serve as the business level aggregation. It is common in many data sets to use identifier fields to define them. These entities are used for roll ups or aggregate metrics for reporting or data modeling purposes. In a fair number of data sets these entities correspond to customers, i.e. people. However, entities need not be constrained to the notion of people. These non-people entities and people based entities share many common traits in the data. You can construct entities for any data concepts that are necessary for the business needs (e.g. retail products, music or organizations). What we really want are entities that best reflect the physical world. Often this requires more than just the identifier fields. The Entity Effect is an entity concept in the data that is realized and can be utilized to further a process through an identified data driven relationship. In this article, we will explore the definition of more complex data entities and provide a simple example to illustrate these ideas.

To view this content
Login OR subscribe for free

Already receive the Predictive Analytics Times emails?
As of January 2014, the Predictive Analytics Times now requires legacy email subscribers to upgrade their subscription - one time only - in order to attain a password-protected login and gain complete access.

Click here to complete this one-time subscription upgrade

  

Existing Users Log In
   
New User Registration
*Required field

Comments are closed.