Logistic regression for business analytics using RapidMiner: Part 2

Posted by Bala Deshpande on Thu, May 12, 2011 @ 08:35 AM

In part 1 we gave a brief introduction to logistic regression and indicated when it might be appropriate to use it in business analytics settings. Probably the best definition of Logistic regression is this " ... a mathematical modeling approach in which the best-fitting, yet least-restrictive model is desired to describe the relationship between several independent explanatory variables and a dependent dichotomous response variable".

Read More

Tags: data mining with rapidminer, business analytics, logistic regression models, logistic regression, lift/gain charts

How to evaluate classification models for business analytics - Part 2

Posted by Bala Deshpande on Tue, May 03, 2011 @ 11:01 AM

In this previous article we described how to construct basic tools such as the "confusion matrix" and Lift/Gain charts for evaluating classification models used for business analytics and predictive analytics. In this article we describe another common evaluation tool - the Receiver Operating Characteristics (ROC) chart and its Area Under Curve (AUC).

Read More

Tags: predictive analytics, business analytics, advanced business analytics, lift/gain charts

How to evaluate classification models for business analytics - Part 1

Posted by Bala Deshpande on Wed, Apr 27, 2011 @ 10:51 AM

The process of developing models for advanced business analytics deployment follows a process that is similar to this diagram.

Read More

Tags: predictive analytics, advanced business analytics, lift/gain charts