关联规则挖掘数据集.txt

apriori经典算法

Data mining is
about finding interesting
structures in data, 
which may be interpreted 
as knowledge about
the data or may be used
to predict events related to the data. 
These structures
take the form of patterns, 
which are concise
descriptions of the data set. 
Data mining makes
the exploration and
exploitation of large
databases easy,
convenient,
and practical
for those who have data
but not years of training
in statistics or data analysis.
The knowledge
extracted by a data mining
algorithm can have many
forms and many uses.
It can be in the form
of a set of rules,
a decision tree,
a regression model,
or a set of associations,
among many other possibilities.
It may be used to produce summaries 
of data or to get insight
into previously unknown correlations.
It also may be used to predict
events related to the data
for example,
missing values,
records for which
some information is not known,
and so forth.
There are many different
data mining techniques,
most of them originating
from the fields of
machine learning,
statistics,
and database programming.
Note Machine learning,
as defined here,
refers to the computer's ability
to improve data mining
algorithms automatically through experience. 
Data training, an important
term that will be used in this context
throughout this specification, 
refers to the process 
where the data mining algorithm
analyzes the input data and finds hidden patterns.
Using this trained data,
these discovered patterns can
then be formed into a model
and applied to the machine's learning process. 
Data mining can be applied for a number of different tasks.
The major ones are predictive
modeling classification,
segmentation clustering,
association,
sequence and deviation analysis,
and dependency modeling. 
This section presents a brief description of each of these tasks.