BUSINESS INTELLIGENCE/BUSINESS ANALYTICS

5.9 BUSINESS INTELLIGENCE/BUSINESS ANALYTICS

Now that we know about databases, data warehouses, data marts, and the analytical decision-making methods discussed in Chapter 4, we are ready to discuss business intelligence/business analytics intelligently.

Business intelligence describes the basic architectural components of a business intelligence environment, ranging from traditional topics, such as business process modeling and data modeling, to more modern topics, such as business rule systems, data profiling, information compliance and data quality, data warehousing, and data mining (see Loshin, 2003).

Business intelligence involves acquiring data and information (and perhaps even knowledge, see Chapter 9) from a wide variety of sources and utilizing them in decision-making. Technically, business analytics adds an additional dimension to busi- ness intelligence: models and solution methods. These are often buried so deep within

273 P A R T 11 DECISION SUPPORT SYSTEMS

sources warehouse

support

Results

Data mining

Visualization Visualization

F I G U R E 5 . 6 TIIR ACTIVITIES O I BUSINESS INTEU ICI.NCF

terms are used interchangeably. We show the activities of business intelligence in Figure 5.6. Business intelligence methods and tools are highly visual in nature. They

provide charts and graphs of multidimensional data with the click of a mouse. These methods generally access data from data warehouses and deposit them into a local,

multidimensional database system. Online analytical processing (OLAP) methods allow an analyst, or even (less typically) a manager to slice and dice the data, while observing graphs and tables that reflect the dimensions being observed. Models may

be applied to the data for forecasting or to identify opportunities (for software exam- ples, see Temtec Executive Viewer, Cognos Impromptu and PowerPlay, and IBM Cube Views). Data mining methods apply statistical and deterministic models, and artificial intelligence methods to data, perhaps guided by an analyst (or manager), to identify hidden relationships or induce/discover knowledge among the various data or text ele- ments (for software examples, see IBM DB2 Intelligent Miner Scoring, Angoss KnowledgeSeeker, Megaputer Intelligence PolyAnalyst, and SAS Enterprise Miner). Data mining is also highly visual in the way results are displayed. Graphs and charts typically display results. Thus the key difference between OLAP and data mining is that data mining runs (mostly) automatically, while OLAP is driven. As tools improve in ease of use, more and more managers utilize them, resulting in a trend to move busi- ness intelligence from the analyst to the user (manager). This introduces a new prob- lem: Managers sometimes do not fully understand business intelligence/business ana- lytics methods. In consequence, their focus may be on visualization rather than application of appropriate and accurate analysis tools. With both tools, it is important to recognize that systems analysts are generally required to set up the access to the data to be analyzed. This involves dealing with data cleansing and integration, a task best left to IS specialists. See the Opening Vignette and DSS in Action 5.7.

All managers and executives should be using business intelligence systems, but some find the data irrelevant or the tools too complicated to use. Sometimes managers are not trained properly. Distributing information from analytics throughout a com- pany is a major challenge; most businesses want a greater percentage of the enterprise to leverage analytics, but most of the challenges around technology involve culture, people, and processes (see Hatcher, 2003). A critical issue is to align BI systems to busi- ness needs. If the system does not provide useful information, it is considered useless. See DSS in Focus 5.24 for details of a recent study on how executives currently utilize

• C H A P T E R 5 DATA WAREHOUSING, ACQUISITION, M I N I N G , BUSINESS ANALYLTICS AND VISUALIZATION

DSS IN FOCUS 5.24