Java Reference
In-Depth Information
Once application owners see the potential of data mining in
these applications unleashed, through the efforts of their application
creators, the application owners may choose to apply more expert
data mining skills to determine whether the quality of the results
can be further enhanced. Oftentimes, even small insights can have a
significant impact on a business or scientific problem.
Introducing Data Mining
Data mining is the process of finding patterns and relationships in
data. At its core, data mining consists of developing a model, which is
typically a compact representation of patterns found using historical
data, and applying that model to new data. We apply a model to data
to predict individual behavior (classification and regression), segment
a population (clustering), determine relationships within a population
(association), as well as to identify the characteristics that most
impact a particular outcome (attribute importance). These and other
data mining capabilities are explored in detail in subsequent chapters.
Data Mining grew as a direct consequence of the availability of large repositories of
data. Data collection in digital form was already underway by the 1960s, allowing
for retrospective data analysis via computers. Relational Databases arose in the
1980s along with Structured Query Languages (SQL), allowing for dynamic, on-
demand analysis of data. The 1990s saw an explosion in growth of data. Data ware-
houses were beginning to be used for storage of data. Data Mining thus arose as a
response to challenges faced by the database community in dealing with massive
amounts of data, application of statistical analysis to data and application of search
techniques from Artificial Intelligence to these problems. [Wikipedia-DM 2006]
Motivations for undertaking data mining include reducing costs,
increasing revenue, making new discoveries, automating laborious
human tasks, identifying fraud, and improving customer or user
experiences. As such, data mining is a competitive strategy for all
industries and ventures.
Data Mining by Other Names
Data mining goes by several aliases, for example, advanced analytics,
predictive analytics, artificial intelligence, and machine learning . Advanced
analytics is commonly thought of as referring to sophisticated statisti-
cal analysis, online analytical processing (OLAP), and data mining. In
the next section, we elaborate on the difference between OLAP and
data mining.
Search WWH ::

Custom Search