Page 395 - Using MIS

Q6 How Do Organizations Use BigData Applications? 363

Figure 9-22
Credit Score Decision Tree
Source: Used with permission of
Insightful Corporation. Copyright ©
1999–2005 Insightful Corporation.
All Rights Reserved.

These two major categories are then further subdivided into three classifications:
CreditScore is a creditworthiness score obtained from a credit agency; MonthsPastDue is the
number of months since a payment; and CurrentLTV is the current ratio of outstanding balance
of the loan to the value of the loan’s collateral.
With a decision tree like this, the financial institution can develop decision rules for accept-
ing or rejecting the offer to purchase loans from another financial institution. For example:
• If percent past due is less than 50 percent, then accept the loan.
• If percent past due is greater than 50 percent and
• If CreditScore is greater than 572.6 and
• If CurrentLTV is less than .94, then accept the loan.
• Otherwise, reject the loan.
Of course, the financial institution will need to combine this risk data with an economic analysis
of the value of each loan to determine which loans to take.
Decision trees are easy to understand and, even better, easy to implement using decision
rules. They also can work with many types of variables, and they deal well with partial data.
Organizations can use decision trees by themselves or combine them with other techniques.
In some cases, organizations use decision trees to select variables that are then used by other
types of data mining tools. For example, decision trees can be used to identify good predictor
variables for neural networks.

Q6 How Do Organizations Use BigData Applications?

As stated in Chapter 5, BigData (also spelled Big Data) is a term used to describe data collections
that are characterized by huge volume, rapid velocity, and great variety. To review:
• BigData data sets are at least a petabyte in size, and usually larger.
• BigData is generated rapidly.
• BigData has structured data, free-form text, log files, possibly graphics, audio, and video.