📄 apriorit_gui.html
字号:
<HTML>
<HEAD>
<TITLE>Apriori-T Association Rule Mining (ARM) Software</TITLE>
</HEAD><BODY BGCOLOR="white">
<CENTER><TABLE BORDER = 0 cellpadding = 10 WIDTH="100%">
<TR><TD BGCOLOR = BB0000>
<CENTER><TABLE BORDER = 0 cellpadding = 10 WIDTH="90%">
<TR><TD BGCOLOR = 006699>
<FONT COLOR = "white">
<CENTER>
<H1>THE LUCS-KDD APRIORI-T ASSOCIATION RULE MINING ALGORITHM</H1>
<HR WIDTH=<50%">
<BR></CENTER>
<TABLE ALIGN=right BGCOLOR="white" BORDER=1 CELLPADDING=5>
<TR><TD>
<img src="../../../GifFolder/logo115.gif"
alt="Liverpool University">
</TD></TABLE>
<P><B><I>Frans Coenen<I></B></P>
<P><B>Department of Computer Science</B></P>
<P><B>The University of Liverpool</B></P>
<P>13 February 2004</P>
</FONT></TD>
</TABLE></CENTER>
</TD></TABLE>
</CENTER>
<BR>
<h2>CONTENTS</h2>
<table BORDER=0 CELLPADDING=0 WIDTH=100%>
<tr><td WIDTH="48%">
<dl>
<dt>1. <a HREF = "#introduction">Introduction</a>.</dt>
<dt>2. <a HREF = "#downloading">Downloading the software</a>.</dt>
<dl>
<dt>2.1. <a HREF = "#compiling">Compiling</A>.</dt>
<dt>2.2. <a HREF = "#documentation">Documentation</A>.</dt>
</DL>
</dl>
</td><td><pre> </pre></td><td>
<dl>
<dt>3. <a HREF = "#running">Running the software</a>.</dt>
</dl>
</td></table><br><hr><br>
<a NAME =" introduction">
<table BORDER-=0 WIDTH="100%" CELLPADDING=5 BGCOLOR="006699">
<tr><td><h2><font color = "white">1. INTRODUCTION</font></h2></td></TABLE>
<P>Apriori-T (Apriori Total) is an Association Rule Mining (ARM) algorithm,
developed by the <A HREF = "http://www.csc.liv.ac.uk/~frans/KDD/">LUCS-KDD
research team</A>. The code obtainable from this page is a GUI version
that includes (for comparison purposes) implementations of Brin's DIC algorithm
(Brin et al. 1997) and Toivonon's negative boarder ARM approach (Toivonen
1996).</P>
<br><hr><br>
<a NAME ="downloading">
<table BORDER-=0 WIDTH="100%" CELLPADDING=5 BGCOLOR="006699">
<tr><td><h2><font color ="white">2. DOWNLOADING THE SOFTWARE</font></h2></td>
</table><BR>
<P>The Apriori-T ARM software comprises four source files. These are provided
from this WWW page together with three application classes. The source files are
as follows:</P>
<OL>
<LI><A HREF = "AprioriT_GUI_App"><TT>AprioriT_GUI_App</TT></A>: Application class.
<LI><A HREF = "AprioriTcontrol"><TT>AprioriTcontrol</TT></A>: Class containing the
GUI control methods.
<LI><A HREF = "AssocRuleMining.java"><TT>AssocRuleMining.java</TT></A>:
Set of general ARM utility methods to allow: (i) data input and input
error checking, (ii) data preprocessing, (iii) manipulation of records (e.g.
operations such as subset, member, union etc.) and (iv) data and parameter
output.
<LI><A HREF = "DIC_Ttree"><TT>DIC_Ttree</TT></A>: Methods to implement the
LUCS-KKD version of the DIC algorithm.
<LI><A HREF = "DIC_TtreeNode"><TT>DIC_TtreeNode</TT></A>: Set of method concerned
with the structure of T-tree nodes adapted for the DIC algorithm.
<LI><A HREF = "RuleNode.java"><TT>RuleNode.java</TT></A>: Set of methods that
allow the creation and manipulation (e.g. ordering, etc.) of a list of ARs.
<LI><A HREF = "TotalSupportTree.java"><TT>TotalSupportTree.java</TT></A>:
Methods to implement the "Apriori-T" algorithm using the "Total support" tree
data structure (T-tree).
<LI><A HREF = "TotalSupportTreeNegBoarder.java"><TT>TotalSupportTreeNegBoarder.java</TT></A>:
Methods to implement a variation of "Apriori-T" algorithm incoporating the negative
boarder concept
<LI><A HREF = "TotalSupportTreeNoXcheck.java"><TT>TotalSupportTreeNoXcheck.java</TT></A>:
Version of Apriori-T algorithm that does not include cross checking. Cross checking
is the process where by the neighbouring branches of the T-tree are checked to ensure
that all the <TT>K-1</TT> subsets of a candidate set are supported (the downward
closure property of itemsets is used in the candidate generation process). Cross checking
carried a time penalty with it and so Apriori-T without cross checking is a faster
algorithm, however it usually requires more storage space as unnecessary candidate sets
are often generated.
<LI><A HREF = "TtreeCanvas"><TT>TtreeCanvas</TT></A>: Methods to create a canvas onto
which the T-tree may be painted if required.
<LI><A HREF = "TtreeNode.java"><TT>TtreeNode.java</TT></A>: Methods concerned
with the structure of T-tree nodes. Arrays of
these structures are used to store nodes at the same level in any sub-branch of
the T-tree. Note this is a separate class to the other T-tree classes which are
arranged in a class hierarchy as illustrated below.
<LI><A HREF = "TtreeWindow"><TT>TtreeWindow</TT></A>: Methods to create a GUI window to
contain a T-tree canvas (if requested).
</OL>
<CENTER>
<TABLE BORDER=1><TR><TD>
<TABLE BORDER=1 CELLPADDING=10><TR><TD>
<CENTER>
<PRE>
AssocRuleMining
|
TotalSupportTree
|
+----------------------+----------------------+
| | |
DIC_Ttree TotalSupportTreeNegBorder TotalSupportTreeNoXcheck
</PRE>
</CENTER>
</TD></TABLE></CENTER>
</TD></TABLE></CENTER>
<BR>
<CENTER>
<TABLE BORDER=1><TR><TD>
<TABLE BORDER=1 CELLPADDING=10><TR><TD>
<CENTER>
<PRE>
TtreeNode
|
DIC_TtreeNode
</PRE>
</CENTER>
</TD></TABLE></CENTER>
</TD></TABLE></CENTER>
<P>There is also a "tar ball" <A HREF = "aprioriT.tgz">aprioriT.tgz</A>
that can be downloaded that includes all of the above source and application
class files. It can be unpacked using <TT>tar -zxf aprioriT.tgz</TT>.</P>
<BR>
<a NAME = "compiling">
<H3>2.1. Compiling</H3>
<p>The software has been implemented in Java which should therefore make it highly
portable. The code does not require any special packages and thus can be
compiled using the standard Java compiler:</p>
<pre>
javac *.java
</pre>
<BR>
<a NAME = "documentation">
<H3>2.2. Documentation</H3>
<P>The code can be documented using <I>Java Doc</I>. First create a directory
<TT>Documentation</TT> in which to place the resulting HTML pages and then type:</P>
<PRE>
javadoc -d Documentation/ *.java
</PRE>
<P>This will produce a hierarchy of WWW pages contained in the <TT>Document</TT>
directory.</P>
<br><hr><br>
<a NAME ="running">
<table BORDER-=0 WIDTH="100%" CELLPADDING=5 BGCOLOR="006699">
<tr><td><h2><font color = "white">3. RUNNING THE SOFTWARE</font></h2></td>
</table><BR>
<p>When compiled the software can be invoked in the normal manner using the
Java interpreter:</p>
<pre>
java AprioriT_GUI_App
</pre>
<p>If you are planning to process a very large data set it is a good idea to
grab some extra memory. For example:</p>
<pre>
java -Xms600m -Xmx600m AprioriT_GUI_App
</pre>
<P>The input to the software, in all cases is a (space separated)
<I>binary valued</I> data set <TT>R</TT>. The set <TT>R</TT>
comprises a set of <TT>N</TT> records such that each record (<TT>r</TT>),
in turn, comprises a set of <I>attributes</I>. Thus:</P>
<PRE>
R = {r | r = subset A}
</PRE>
<P>Where <TT>A</TT> is the set of available attributes. The value <TT>D</TT> is
then defined as:
<PRE>
D = |A|
</PRE>
<P>We then say that a particular data set has <TT>D</TT> <I>columns</I> and
<TT>N</TT> <I>rows</I>. A small example data sets might be as follows:</P>
<PRE>
1 2 3 6
1 4 5 7
1 3 4 6
1 2 6
1 2 3 4 5 7
</PRE>
<P>where, in this case, <TT>A = {1, 2, 3, 4, 5, 6, 7}</TT>. Note that
attribute numbers are ordered sequentially commencing with the
number 1 (the value 0 has a special meaning).</P>
<br><hr><br>
<table BORDER-=0 WIDTH="100%" CELLPADDING=5 BGCOLOR="006699">
<tr><td><h2><font color =" white">REFERENCES</font></h2></td>
</table> <BR>
<DL>
<DT>Brin, S., Motwani, R., Ullman, J.D. and Tsur S. (1997).</DT>
<DD><I>Dynamic itemset counting and implication rules for market basket data.</I>
SIGMOD Record, Vol 6, No 2, ACM Press, pp 255-264.</DD>
<DT>Toivonen, H. (1996)</DT>
<DD><I>Sampling Large Databases for Association Rules</I>.
Proc. 1996 Int. Conf. Very Large Data Bases, pp. 134-145.</DD>
</DL>
<br><hr><br>
<p>Created and maintained by
<a HREF=http://www.csc.liv.ac.uk/~frans/>Frans Coenen</a>.
Last updated 13 March 2007</P>
</body>
</html>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -