📄 page_97.html
字号:
<HTML> <HEAD> <!--SCRIPT LANGUAGE="JavaScript" SRC="http://a1835.g.akamai.net/f/1835/276/3h/www.netlibrary.com/include/js/dictionary_library.js"></SCRIPT> <SCRIPT LANGUAGE="JavaScript"> if (!opener){document.onkeyup=parent.turnBookPage;} </SCRIPT!--> <META HTTP-EQUIV="Cache-Control" CONTENT="no-cache"> <META HTTP-EQUIV="Pragma" CONTENT="no-cache"> <META HTTP-EQUIV="Expires" CONTENT="-1"><META http-equiv="Content-Type" content="text/html; charset=windows-1252"><SCRIPT>var PrevPage="Page_96";var NextPage="Page_98";var CurPage="Page_97";var PageOrder="108";</SCRIPT> <TITLE>Document</TITLE> </HEAD> <BODY BGCOLOR="#FFFFFF"><CENTER><TABLE BORDER=0 WIDTH=100% CELLPADDING=0><TR><TD ALIGN=CENTER> <TABLE BORDER=0 CELLPADDING=2 CELLSPACING=0 WIDTH=100%> <TR> <TD ALIGN=LEFT><A HREF='Page_96.html'>Previous</A></TD> <TD ALIGN=RIGHT><A HREF='Page_98.html'>Next</A></TD> </TR> </TABLE></TD></TR><TR><TD ALIGN=LEFT><P><A NAME='JUMPDEST_Page_97'/><A NAME='{354}'/><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=RIGHT><FONT FACE='Times New Roman, Times, Serif' SIZE=2 COLOR=#FF0000>Page 97</FONT></TD></TR></TABLE><A NAME='{355}'/><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0><TR> <TD ROWSPAN=5></TD> <TD COLSPAN=3 HEIGHT=12></TD> <TD ROWSPAN=5></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR><TD></TD> <TD><FONT FACE='Times New Roman, Times, Serif' SIZE=3>Keep in mind that neural networks are basically memories: they do not compute solutions as much as remember answers. As such, the data samples that are used to train a network are critical in determining the accuracy of the network model. If the memory is organized in the proper way, then the look-up table of that network and its core weights, which is really what networks are all about, will generalize in a manner which will be very accurate. Despite the rap networks take for being black-box wonders from which no explanation to their solutions is forthcoming (which, ironically, may be said for regression), they are very good tools. Neural networks are good data mining tools for classification and prediction and work best when used in conjunction with machine-learning and genetic algorithms.</FONT></TD><TD></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR> <TD COLSPAN=3 HEIGHT=1></TD></TR></TABLE><A NAME='{356}'/><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0><TR> <TD ROWSPAN=5></TD> <TD COLSPAN=3 HEIGHT=17></TD> <TD ROWSPAN=5></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR><TD></TD> <TD><FONT FACE='Times New Roman, Times, Serif' SIZE=3><B>Neural Network Tools</B></FONT></TD><TD></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR> <TD COLSPAN=3 HEIGHT=1></TD></TR></TABLE><A NAME='{357}'/><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0><TR> <TD ROWSPAN=5></TD> <TD COLSPAN=3 HEIGHT=12></TD> <TD ROWSPAN=5></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR><TD></TD> <TD><FONT FACE='Times New Roman, Times, Serif' SIZE=3>Most of the work involved in the use of neural networks is in the preparation of the data. Luckily, most of today's data mining tools are sophisticated enough to handle the dirty work involved in such tasks as balancing the data. (<I>Balancing</I> is using an equal sample of positive and negative cases (Buyers vs. Non-Buyers) to train a network.) Another task involves the scaling of the data; converting all inputs to a range from 0 to 1 and back in the output. This may also involve using functions for dealing with skewed data sets, such as taking the logarithm or the square root of their values prior to passing the data through a network. Still another task involves converting categorical inputs into either 1-of-N values or thermometer values:</FONT></TD><TD></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR> <TD COLSPAN=3 HEIGHT=1></TD></TR></TABLE><A NAME='{358}'/><TABLE CELLPADDING=0 CELLSPACING=0 BORDER=0 WIDTH='100%'><TR><TD HEIGHT=12></TD></TR><TR><TD><TABLE CELLSPACING=0 WIDTH=583 CELLPADDING=5><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2><B>If Number of Children is:</B></FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2><B>1-of-N Value</B></FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2><B>Thermometer Value</B></FONT></TD></TR></TABLE></TD></TR><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>0</FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>00001</FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>0.5000</FONT></TD></TR></TABLE></TD></TR><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>1</FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>00010</FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>0.7500</FONT></TD></TR></TABLE></TD></TR><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>2</FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>00100</FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>0.8750</FONT></TD></TR></TABLE></TD></TR><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>3</FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>01000</FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>0.9375</FONT></TD></TR></TABLE></TD></TR><TR><TD WIDTH=186 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>4</FONT></TD></TR></TABLE></TD><TD WIDTH=205 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>10000</FONT></TD></TR></TABLE></TD><TD WIDTH=192 VALIGN=TOP><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0 WIDTH='100%'><TR><TD ALIGN=CENTER><FONT FACE='Times New Roman, Times, Serif' SIZE=2>1.0000</FONT></TD></TR></TABLE></TD></TR></TABLE></TD></TR></TABLE><BR><A NAME='{359}'/><TABLE BORDER=0 CELLSPACING=0 CELLPADDING=0><TR> <TD ROWSPAN=5></TD> <TD COLSPAN=3 HEIGHT=12></TD> <TD ROWSPAN=5></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR><TD></TD> <TD><FONT FACE='Times New Roman, Times, Serif' SIZE=3>The following data mining tools incorporate back propagation and, in some instances, SOM networks for classification. Most of these tools are mature, third- or fourth-generation versions of their software, which provide some functions for data preparation and balancing of inputs prior to training and model creation. Care must still</FONT><FONT FACE='Times New Roman, Times, Serif' SIZE=3 COLOR=#FFFF00><!-- continue --></FONT></TD><TD></TD></TR><TR> <TD COLSPAN=3></TD></TR><TR> <TD COLSPAN=3 HEIGHT=1></TD></TR></TABLE><A NAME='{35A}'/></FORM></P></TD></TR></TABLE><P><FONT SIZE=0 COLOR=WHITE></CENTER><A NAME="bottom"> </A><!-- netLibrary.com Copyright Notice --> </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 + -