Togaware DATA MINING
Desktop Survival Guide
by Graham Williams
Google


Temporal Analysis

A common data mining task with temporal data is to find repeating patterns in the data - see frequent closed itemsets.

SNN Clustering () and ().

A common question is how likely an event will occur at a give point in time. Suppose we had some simple churn data recording how long a customer has been with a telecoms provider before they churned.



ID   Gender  Months     Churn
1       M        12         1
2       M         5         0
3       M        32         1
4       M         4         0
5       M        10         1
6       F        12         0
7       F         5         1
8       F        15         0
9       F         5         1
10      F        12         0

We may be tempted in the first instance to us a logistic regression including Gender and Months to predict Churn with:


\begin{displaymath}
logit(Churn=1)=b_0+b_1*Gender+b_2*Tenure
\end{displaymath}



Copyright © 2004-2010 Togaware Pty Ltd
Support further development through the purchase of the PDF version of the book.
The PDF version is a formatted comprehensive draft book (with over 800 pages).
Brought to you by Togaware. This page generated: Sunday, 22 August 2010