Filter
Top Products
Chapter
4
44
4
Understanding Data Mining
Data Mining Overview
Through a variety of techniques, data mining identifies nuggets of information in bodies of data.
Data mining extracts information in such a way that it can be used in areas such as decision
support, prediction, forecasts, a nd estimation. Data is often vol
uminous but of low v alue and with
little direct usefulness in its raw form. It is the hidden information in the data that has value.
In data mining, success comes from combin ing your (or your expert’s) knowledge of the
data with advanced, active analysis techniques in whic h the compu
ter identifies the underlying
relationships an d features in the data. The process of data mining generates models from historical
data that are later used for predictions, pattern detection, and more. The technique for building
these models is called machine learning or modeling.
Modeling Techniques
IBM® SPSS® Modeler includes a number of machine-lea r ning and modeling technologies, which
can be roughly grouped according to the types of problems they are intended to solve.
Predictive modelin g methods include decisi on trees, neural networks, and statistical models.
Cluste r ing models focus on identifying groups of similar records and labeling the records
according to the group to which they belong. Clusterin g methods include Kohonen, k-means,
and TwoStep.
Associa tion rules associate a particula r conclusion (such as the purchase of a particular
product) with a set of conditions (the purchase of seve r al other products).
Screening m odels c an be used to screen data to locate fields and r ecords that are most likely to
be of interest in modeling and identify outliers that may not fit known patterns. Available
methods include feature selection and anomaly detection.
Data Manipulation and Discovery
SPSS Modeler also includes many facilities that let you apply your expe r tise to the data:
Data manipulation.
Constructs new data items derived from existing ones and breaks down the
data into meaningful subset s. Data from a variety o f sources can be me rged and filtered.
Browsing and visualization.
Displays aspects of the data using the Data Audit node to perform
an initial a udit including graphs and statistics. Advanced visualization includes interactive
graphics, which can be exported for inclusio n in project repor ts .
Statistics.
Confirms suspected relationships between variables in the data. Statistics from
IBM® SPSS® Statistics can also be used within SPSS Modeler.
Hypothesis testing.
Constructs models of how the data behaves and verifie s these mo dels.
© Copyright IBM Corporation 1994, 2012.
29