Page 168 - DCAP208_Management Support Systems

Unit 10: Data Mining Tools and Techniques

Notes
Table 10.2: Some of the Differences between the Nearest-Neighbor Data Mining
Technique and Clustering
Nearest Neighbor Clustering
Used for prediction as well as Used mostly for consolidating data into a high-level
consolidation. view and general grouping of records into like
behaviors.
Space is defined by the problem to be Space is defined as default n-dimensional space, or
solved (supervised learning). is defined by the user, or is a predefined space
driven by past experience (unsupervised learning).
Generally only uses distance metrics to Can use other metrics besides distance to determine
determine nearness. nearness of two records - for example linking two
points together.
10.2.4 Decision Trees

A decision tree is a predictive model that, as its name implies, can be viewed as a tree. Specifically
each branch of the tree is a classification question and the leaves of the tree are partitions of the
dataset with their classification.
For example, if we were going to classify customers who churn (don’t renew their phone contracts)
in the Cellular Telephone Industry a decision tree might look something like that found in
Figure 10.2.
Figure 10.2: A Decision Tree is a Predictive Model that makes a Prediction on the Basis of
a Series of Decision much Like the Game of 20 Questions

50 Churners
50 Non-Churners
New Technology Phone?

Yes No
30 Churners 20 Churners
50 Non-Churners 0 Non-Churners
Customer < 2.3 years?
Yes No
25 Churners 5 Churners
10 Non-Churners 40 Non-Churners
Age < 55
Yes No
20 Churners 5 Churners
0 Non-Churners 10 Non-Churners

You may notice some interesting things about the tree:

It divides up the data on each branch point without losing any of the data (the number of
total records in a given parent node is equal to the sum of the records contained in its two
children).

The number of churners and non-churners is conserved as you move up or down the tree
It is pretty easy to understand how the model is being built (in contrast to the models from
neural networks or from standard statistics).

LOVELY PROFESSIONAL UNIVERSITY 161