Game Development Reference
In-Depth Information
Artificial Intelligence and
Intelligent Business
and Pugh, 1981). With the emergence of the Dy-
namic Decision-Making in Systems Dynamics,
simulators integrate this complexity, consider-
ing that decision makers do not have consistent
and persistent goal structures, they are not using
complex algorithms or calculating the optimal
solution (there are no optimal solutions in the
economy), nor are they trying to get all information
to support decision making; not all decisions are
rational, and non-linear quantitative processes are
not developed (Simon, 1955), (Hogarth, 1980),
(Döner, 1980).
On the other hand, these aspects have arisen
in discussions about the type of business game
simulator to use: “black box” models or “transpar-
ent information” models. The black box model is
justified by the arguments stated above: decision
makers do not use all the information, they do not
make calculations, nor do they solve complex
problems, and are not entirely rational. Difficult
interrelations and non-linear methodologies are
difficult to interpret by the user, and it is not clear
whether they are worthy for education. Business
Games with transparent information, white box,
or transparent box, offer the user a tool of causal
diagrams and provide the methodology for making
calculations, but have the disadvantage that they
can only use equations and simple relationships
since complex relationships and mathematical
models would not be understandable for the user
(Kemeny and Kreutzer 1992) (Isaacs and Senge,
1992) (Musselwhite, 2006). Although cause-ef-
fect relations and the knowledge of algorithms
improve their understanding and assimilation,
technical reductions and simplification excesses
reduce the quality of the results obtained by the
management simulator, which after all, aim to
simulate the reality of the company in competition,
with a complex contingent environment.
The origin of Business Intelligence (BI) can
be traced back to the first data processing ap-
plications (McDonald and Wilmsmeier, 2004).
Currently, business organizations are moving
towards decision-making processes that are based
on information. Business Intelligence represents
technologies and methods for following the best
strategy in the marketplace. Artificial Intelligence
includes several technologies that may be very
useful in improving BI, such as Data Mining,
Evolutionary Computation and Intelligent Agents
(Russel and Norvig, 1999).
Adaptive business intelligence combines
optimization and prediction techniques to help
decision makers to take the best decisions in com-
plex and rapidly changing environments. In these
systems there are two main questions: What will
happen in the future? And what is the best action
to take? In these systems, Artificial Intelligence
techniques such as decision trees, artificial neural
networks, agent-based modeling are used.
Knowledge management technologies are less
mature than BI technologies, but these techniques
are now combining today's content management
systems and the Web with improved searching
and data mining capabilities. This combination
is usually called BIKM (Business Intelligence
and Knowledge Management) (Cody, Kreulen,
Krishna and Spangler, 2002).
Data mining techniques are frequently used to
study and to discover concepts and relations in
database marketing (Santos, Cortez, Quintela, and
Pinto 2005). In this case, the usual objective is to
find a set of rules that explain clusters of clients
with homogeneous behaviors. In this work, the
authors validate and eliminate irrelevant data and
use Self-Organizing Maps (SOM) to explore the
clustering space. Finally, decision trees are used
in order to obtain the perfect set of classification
rules.
Search Nedrilad ::




Custom Search