Big Data, Data Warehouses, and Business Intelligence Systems - Database Processing: Fundamentals, Design, and Implementation - page 537

BI systems process data that are extracted from operational databases. In this situation, they
manage the extracted database using a BI DBMS, which may be the same as or different from
the operational DBMS. Finally, BI systems read data purchased from data vendors.

Reporting Systems and Data Mining Applications

BI systems fall into two broad categories: reporting systems and data mining applications.
Reporting systems sort, filter, group, and make elementary calculations on operational data.
Data mining applications , in contrast, perform sophisticated analyses on data, analyses that
usually involve complex statistical and mathematical processing. The characteristics of BI
applications are summarized in Figure 12-3.

Reporting Systems
Reporting systems filter, sort, group, and make simple calculations. All reporting analyses can
be performed using standard SQL, though extensions to SQL, such as those used for OnLine
Analytical Processing (OLAP) , are sometimes used to ease the task of report production.
Reporting systems summarize the current status of business activities and compare that
status with past or predicted future activities. Report delivery is crucial. Reports must be deliv-
ered to the proper users on a timely basis, in the appropriate format. For example, reports may
be delivered on paper, via a Web browser, or in some other format.

Data Mining Applications
Data mining applications use sophisticated statistical and mathematical techniques to per-
form what-if analyses, to make predictions, and to facilitate decision making. For example,
data mining techniques can analyze past cell phone usage and predict which customers are
likely to switch to a competing phone company. Or data mining can be used to analyze past
loan behavior to determine which customers are most (or least) likely to default on a loan.
Report delivery is not as important for data mining systems as it is for reporting systems.
First, most data mining applications have only a few users, and those users have sophisticated
computer skills. Second, the results of a data mining analysis are usually incorporated into
some other report, analysis, or information system. In the case of cell phone usage, the charac-
teristics of customers who are in danger of switching to another company may be given to the
sales department for action. Or the parameters of an equation for determining the likelihood
of a loan default may be incorporated into a loan approval application.

Figure 12-3
Characteristics of Business
Intelligence Applications

• Reporting
- Filter, sort, group, and make simple calculations
- Summarize current status
- Compare current status to past or predicted status
- Classify entities (customers, products, employees, etc.)
- Report delivery crucial
• Data Mining

- Often employ sophisticated statistical and mathematical

techniques

- Used for:

• What-if analyses

• Predictions

• Decisions

- Results often incorporated into some other report

or system