SQL/MX Data Mining Guide

Introduction
HP NonStop SQL/MX Data Mining Guide—523737-001
1-4
Defining the Business Opportunity
6. Create the data mining view.
Transform the data into a mining view, a form in which all attributes about the
primary mining entity occur in a single record.
See Creating the Mining View on page 1-10.
7. Mine the data and build models.
Core knowledge discovery techniques are applied to gain insight, learn patterns, or
verify hypotheses. The main tasks are either predictive or descriptive in nature.
Predictive tasks involve trying to determine what will happen in the future, based
upon historical data. Descriptive tasks involve finding patterns describing the data.
See Mining the Data on page 1-10.
8. Deploy models.
Deployment can take many different forms. For example, deployment might be as
simple as documenting and reporting the results, or deployment might be
embedding the model in an operational system to achieve predictive results.
9. Monitor model performance.
Performance of the model must be monitored for accuracy. When accuracy begins
to decline, the model must be updated to fit the current situation.
See Knowledge Deployment and Monitoring on page 1-11.
In Step 1, a business opportunity is identified and defined. In Steps 2 through 6, data
mining data is gathered, preprocessed, and organized in a form that is suitable for
mining. These steps require the most time in the process. For example, selecting the
data is an important step in the process and typically requires the assistance of a data
mining expert or subject matter expert who has knowledge of the data to be mined.
In Step 7, models are built. In Steps 8 and 9, the models are deployed and monitored.
This latter part of the knowledge discovery process focuses on analyzing the data
mining view prepared in Steps 2 through 6.
Defining the Business Opportunity
The process begins with the identification and precise specification of a business
opportunity. Several factors must be considered when evaluating potential
opportunities:
•
Quantification of the return on investment
What is the answer worth? How much money can be saved? How much of a
competitive advantage does it offer?