Multi-DimensionalDataVisualizationToolforKnowledge

  • PUDN用户
    了解作者
  • matlab
    开发工具
  • 641KB
    文件大小
  • rar
    文件格式
  • 0
    收藏次数
  • 1 积分
    下载积分
  • 11
    下载次数
  • 2009-11-22 16:32
    上传日期
一个多维数据挖掘工具只是挖掘数据库 数据挖掘方向论文
Multi-DimensionalDataVisualizationToolforKnowledge
  • 一个多维数据挖掘工具只是挖掘数据库.pdf
    710.5KB
内容介绍
<html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta charset="utf-8"> <meta name="generator" content="pdf2htmlEX"> <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <link rel="stylesheet" href="https://static.pudn.com/base/css/base.min.css"> <link rel="stylesheet" href="https://static.pudn.com/base/css/fancy.min.css"> <link rel="stylesheet" href="https://static.pudn.com/prod/directory_preview_static/6260f84adaa0590bf7be5bf5/raw.css"> <script src="https://static.pudn.com/base/js/compatibility.min.js"></script> <script src="https://static.pudn.com/base/js/pdf2htmlEX.min.js"></script> <script> try{ pdf2htmlEX.defaultViewer = new pdf2htmlEX.Viewer({}); }catch(e){} </script> <title></title> </head> <body> <div id="sidebar" style="display: none"> <div id="outline"> </div> </div> <div id="pf1" class="pf w0 h0" data-page-no="1"><div class="pc pc1 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://static.pudn.com/prod/directory_preview_static/6260f84adaa0590bf7be5bf5/bg1.jpg"><div class="t m0 x1 h2 y1 ff1 fs0 fc0 sc0 ls0 ws0">A </div><div class="t m1 x2 h2 y1 ff1 fs0 fc0 sc0 ls1 ws0">Multi-Dimensional <span class="_ _0"></span>Data Visualization <span class="_ _1"></span><span class="ls2">Tool </span></div><div class="t m2 x3 h3 y1 ff1 fs1 fc0 sc0 ls0 ws0">for </div><div class="t m1 x4 h2 y2 ff1 fs0 fc0 sc0 ls1 ws0">Knowledge <span class="_ _1"></span>Discovery <span class="_ _2"></span><span class="ls0">in <span class="_ _2"></span><span class="ls1">Databases </span></span></div><div class="t m3 x5 h4 y3 ff2 fs2 fc0 sc0 ls0 ws0">Hing-Yan <span class="_ _2"></span>Lee, Hwee-Leng <span class="_ _2"></span>Ong, Eng-Whatt <span class="_ _3"></span><span class="ls3">Toh, <span class="_ _3"></span><span class="ls4">and <span class="ls0">Sieu-Kong </span>Chan </span></span></div><div class="t m3 x1 h4 y4 ff2 fs2 fc0 sc0 ls0 ws0">Information <span class="_ _2"></span>Technology <span class="_ _2"></span>Institute, National Computer <span class="_ _3"></span>Board </div><div class="t m4 x6 h4 y5 ff2 fs2 fc0 sc0 ls0 ws0">7 </div><div class="t m5 x7 h5 y5 ff2 fs3 fc0 sc0 ls0 ws0">1 </div><div class="t m3 x8 h4 y5 ff2 fs2 fc0 sc0 ls0 ws0">Science </div><div class="t m6 x9 h6 y5 ff1 fs4 fc0 sc0 ls5 ws0">Park </div><div class="t m3 xa h4 y5 ff2 fs2 fc0 sc0 ls0 ws0">Drive, <span class="_ _4"></span>Singapore </div><div class="t m7 xb h5 y5 ff2 fs3 fc0 sc0 ls6 ws0">05 </div><div class="t m3 xc h4 y5 ff2 fs2 fc0 sc0 ls0 ws0">1 <span class="_ _5"></span>1 </div><div class="t m3 xd h4 y6 ff2 fs2 fc0 sc0 ls0 ws0">Republic </div><div class="t m8 xe h7 y7 ff2 fs5 fc0 sc0 ls7 ws0">of </div><div class="t m3 xf h4 y7 ff2 fs2 fc0 sc0 ls0 ws0">Singapore </div><div class="t m3 x10 h4 y8 ff2 fs2 fc0 sc0 ls0 ws0">email: </div><div class="t m9 x11 h8 y9 ff3 fs6 fc0 sc0 ls0 ws0">{ </div><div class="t m3 x12 h4 y9 ff2 fs2 fc0 sc0 ls0 ws0">hingyan, <span class="_ _2"></span>hweeleng, <span class="_ _1"></span>engwhatt, <span class="_ _2"></span>sieukong} <span class="_ _6"></span>@iti.gov.sg </div><div class="t ma x13 h9 ya ff1 fs7 fc0 sc0 ls8 ws0">Abstract </div><div class="t mb x14 ha yb ff4 fs8 fc0 sc0 ls9 ws0">This <span class="lsa">paper <span class="_ _4"></span></span>describes <span class="_ _7"> </span><span class="lsb">a <span class="_ _8"></span><span class="lsc">multi-dimensional <span class="_ _9"> </span>visualization </span></span></div><div class="t mb x14 ha yc ff4 fs8 fc0 sc0 lsd ws0">(MDV) <span class="_ _2"></span><span class="ls9">technique, <span class="_"> </span><span class="lsb">a <span class="_"> </span><span class="lsc">refined <span class="_ _a"> </span>and <span class="_ _9"> </span>improved <span class="_ _7"> </span><span class="lse">form </span></span></span></span></div><div class="t mc x15 hb yd ff5 fs9 fc0 sc0 lsf ws0">of </div><div class="t mb x16 ha yd ff4 fs8 fc0 sc0 lsb ws0">a </div><div class="t mb x14 ha ye ff4 fs8 fc0 sc0 lsc ws0">computational <span class="_ _b"> </span>geometry <span class="_ _b"> </span>method <span class="_ _b"> </span>known <span class="_ _c"> </span><span class="lsa">as <span class="_ _9"> </span><span class="ls9">parallel </span></span></div><div class="t mb x14 ha yf ff4 fs8 fc0 sc0 lsc ws0">coordinates, <span class="_ _8"></span><span class="ls9">that <span class="_ _4"></span><span class="lsb">has been </span></span>realized </div><div class="t m4 x6 hc y10 ff4 fs9 fc0 sc0 ls10 ws0">by </div><div class="t mb x17 ha y10 ff4 fs8 fc0 sc0 ls9 ws0">WinViz. </div><div class="t md x18 hd y10 ff4 fsa fc0 sc0 ls11 ws0">It </div><div class="t mb x19 ha y10 ff4 fs8 fc0 sc0 ls0 ws0">shows </div><div class="t mb x14 ha y11 ff4 fs8 fc0 sc0 lsb ws0">how <span class="_ _2"></span><span class="ls9">MDV <span class="_ _d"></span>differs <span class="_ _3"></span><span class="lsc">significantly <span class="_ _3"></span><span class="lse">from <span class="_ _0"></span><span class="ls9">parallel <span class="_ _7"></span><span class="lsc">coordinates </span></span></span></span></span></div><div class="t mb x14 ha y12 ff4 fs8 fc0 sc0 lsa ws0">as <span class="ls9">originally <span class="_ _4"></span><span class="lsc">conceived <span class="_ _e"> </span></span></span>by <span class="_ _2"></span><span class="lsc">Alfred <span class="_ _e"> </span>Inselberg. <span class="_ _c"> </span><span class="lsb">The <span class="_ _2"></span><span class="lsa">paper </span></span></span></div><div class="t mb x14 ha y13 ff4 fs8 fc0 sc0 ls9 ws0">also <span class="_ _f"> </span>discusses <span class="_ _10"> </span><span class="lsb">WinViz's <span class="_ _11"> </span><span class="lsc">underlying <span class="_ _12"> </span>design </span></span></div><div class="t mb x14 ha y14 ff4 fs8 fc0 sc0 lsc ws0">considerations. <span class="_ _13"> </span><span class="lsb">A <span class="_ _14"> </span>primary <span class="_ _15"> </span><span class="ls9">principle <span class="_ _b"> </span></span></span>is <span class="_ _c"> </span>known <span class="_ _c"> </span>as </div><div class="t mb x1a ha y15 ff4 fs8 fc0 sc0 lsc ws0">"seeing <span class="_"> </span><span class="lsb">your <span class="_"> </span>data <span class="_ _16"> </span><span class="lsa">in <span class="_ _e"> </span></span>a <span class="_ _16"> </span></span>single <span class="_ _16"> </span><span class="ls9">picture." <span class="_ _17"> </span></span>The <span class="_ _16"> </span>novel </div><div class="t mb x14 ha y16 ff4 fs8 fc0 sc0 lsc ws0">visualization <span class="_ _18"> </span><span class="ls9">intelface <span class="_ _c"> </span></span>of <span class="_ _c"> </span><span class="lsb">WinViz <span class="_"> </span></span>supports <span class="_ _c"> </span>both <span class="_ _c"> </span>the </div><div class="t mb x14 ha y17 ff4 fs8 fc0 sc0 ls9 ws0">display </div><div class="t me x1b he y18 ff6 fsb fc0 sc0 ls12 ws0">of </div><div class="t mb x1c ha y18 ff4 fs8 fc0 sc0 lsc ws0">multi-dimensional <span class="_ _19"> </span>data <span class="_ _1a"> </span><span class="ls0">and <span class="_ _1b"> </span></span>the <span class="_ _1a"> </span>visual </div><div class="t mb x1d ha y19 ff4 fs8 fc0 sc0 lsa ws0">formulation </div><div class="t mf x1c hf y1a ff6 fsc fc0 sc0 ls13 ws0">of </div><div class="t mb x1e ha y1a ff4 fs8 fc0 sc0 lsb ws0">query <span class="_ _e"> </span><span class="ls9">interactively. </span></div><div class="t m10 x1f h10 y1a ff4 fsb fc0 sc0 ls14 ws0">In </div><div class="t mb x20 ha y1a ff4 fs8 fc0 sc0 ls9 ws0">particular, <span class="_ _a"> </span><span class="ls15">we </span></div><div class="t mb x14 ha y1b ff4 fs8 fc0 sc0 lsc ws0">compare <span class="_ _e"> </span>WinViz's <span class="lsb">query </span>operators <span class="_ _7"> </span>with <span class="_ _d"></span><span class="ls9">those </span></div><div class="t m11 x21 h11 y1c ff4 fsd fc0 sc0 ls16 ws0">in </div><div class="t m12 x22 h10 y1c ff4 fsb fc0 sc0 ls17 ws0">SQL </div><div class="t m13 x23 hd y1c ff4 fsa fc0 sc0 ls18 ws0">to </div><div class="t mb x14 ha y1d ff4 fs8 fc0 sc0 lsc ws0">highlight <span class="_ _7"> </span><span class="ls9">its <span class="_ _4"></span>salient <span class="_ _2"></span><span class="lsa">features <span class="_ _8"></span><span class="lsc">realized <span class="_ _7"> </span><span class="lsb">through <span class="_ _4"></span></span>the <span class="_ _8"></span>MDV </span></span></span></div><div class="t mb x14 ha y1e ff4 fs8 fc0 sc0 ls9 ws0">intelface. <span class="_ _b"> </span><span class="lsb">Applications </span></div><div class="t me x24 he y1f ff6 fsb fc0 sc0 ls12 ws0">of </div><div class="t mb x25 ha y1f ff4 fs8 fc0 sc0 lsb ws0">WinViz <span class="_ _4"></span><span class="lsc">include <span class="_"> </span>exploratory </span></div><div class="t mb x14 ha y20 ff4 fs8 fc0 sc0 lsb ws0">data <span class="_ _3"></span><span class="ls9">analysis <span class="lsc">and <span class="_ _d"></span>knowledge <span class="_ _7"> </span><span class="lsb">discovery </span></span></span></div><div class="t m14 x26 h12 y21 ff5 fse fc0 sc0 ls19 ws0">of </div><div class="t mb x27 ha y21 ff4 fs8 fc0 sc0 lsc ws0">databases <span class="ls1a">for </span></div><div class="t mb x14 ha y22 ff4 fs8 fc0 sc0 lsb ws0">a <span class="_ _3"></span><span class="lsc">diverse <span class="_ _8"></span>range </span></div><div class="t m15 x28 hb y23 ff6 fs9 fc0 sc0 ls1b ws0">of </div><div class="t mb x29 ha y23 ff4 fs8 fc0 sc0 lsc ws0">domains. </div><div class="t m16 x14 h13 y24 ff1 fsd fc0 sc0 ls0 ws0">Keywords: </div><div class="t m12 x2a h14 y24 ff2 fs9 fc0 sc0 ls0 ws0">AI <span class="_ _16"> </span>tool, <span class="_"> </span><span class="ls1c">data <span class="_ _e"> </span></span>visualization, <span class="_ _9"> </span><span class="ls1d">data <span class="_"> </span>mining, </span></div><div class="t m12 x14 h14 y25 ff2 fs9 fc0 sc0 ls0 ws0">knowledge <span class="_ _14"> </span>discovery <span class="_ _14"> </span><span class="ls1d">in <span class="_ _1c"> </span></span>databases, <span class="_ _18"> </span>intelligent <span class="_ _1b"> </span><span class="ls1e">user </span></div><div class="t m12 x14 h14 y26 ff2 fs9 fc0 sc0 ls0 ws0">interfaces </div><div class="t m4 x1a h15 y27 ff1 fs2 fc0 sc0 ls1f ws0">1. </div><div class="t ma x2b h9 y27 ff1 fs7 fc0 sc0 ls0 ws0">Introduction </div><div class="t m12 x2c h14 y28 ff2 fs9 fc0 sc0 ls0 ws0">With <span class="_"> </span><span class="ls1d">the <span class="_ _e"> </span></span>pervasive <span class="_ _e"> </span>spread </div><div class="t m17 x12 h16 y28 ff2 fsf fc0 sc0 ls20 ws0">of </div><div class="t m12 x2d h14 y28 ff2 fs9 fc0 sc0 ls0 ws0">computers <span class="_"> </span>in </div><div class="t m18 x2e h16 y28 ff2 fsf fc0 sc0 ls21 ws0">every </div><div class="t m12 x14 h14 y29 ff2 fs9 fc0 sc0 ls0 ws0">facet </div><div class="t m19 x2f h17 y29 ff2 fs10 fc0 sc0 ls22 ws0">of </div><div class="t m12 x30 h14 y29 ff2 fs9 fc0 sc0 ls0 ws0">business <span class="_ _7"> </span><span class="ls1d">and <span class="_ _7"> </span></span>industry, <span class="_ _e"> </span><span class="ls1d">data <span class="_ _8"></span>is <span class="_ _e"> </span></span>being <span class="_ _7"> </span>routinely </div><div class="t m12 x14 h14 y2a ff2 fs9 fc0 sc0 ls0 ws0">and <span class="_ _9"> </span>almost <span class="_ _9"> </span>mindlessly <span class="_ _a"> </span>captured. <span class="_ _1d"> </span>Often <span class="_ _9"> </span><span class="ls1e">beyond <span class="_ _16"> </span></span>the </div><div class="t m12 x14 h14 y2b ff2 fs9 fc0 sc0 ls1d ws0">current <span class="_ _4"></span><span class="ls0">window <span class="_ _3"></span><span class="ls23">of <span class="_"> </span><span class="ls0">interest <span class="_ _d"></span><span class="ls1d">(e.g., </span>to produce this <span class="_ _d"></span>month's </span></span></span></div><div class="t m12 x14 h14 y2c ff2 fs9 fc0 sc0 ls0 ws0">bank <span class="_"> </span>statement), <span class="_"> </span><span class="ls1d">the <span class="_ _e"> </span>data <span class="_ _e"> </span></span>collected <span class="_ _7"> </span><span class="ls1d">is <span class="_"> </span></span>archived <span class="_"> </span>away, </div><div class="t m12 x14 h14 y2d ff2 fs9 fc0 sc0 ls0 ws0">often <span class="_ _8"></span>never <span class="_ _d"></span><span class="ls1e">to <span class="_ _7"> </span><span class="ls24">be <span class="_ _7"> </span></span>used <span class="_ _7"> </span><span class="ls1d">again. <span class="_ _18"> </span></span></span>Knowledge <span class="_ _d"></span>discovery <span class="ls1d">in </span></div><div class="t m12 x14 h14 y2e ff2 fs9 fc0 sc0 ls0 ws0">databases <span class="_ _15"> </span><span class="ls1d">(KDD), <span class="_ _16"> </span></span>also <span class="_ _16"> </span>known <span class="_ _14"> </span>as <span class="_ _9"> </span><span class="ls1d">data <span class="_ _9"> </span>mining, <span class="_ _15"> </span></span>has </div><div class="t m12 x14 h14 y2f ff2 fs9 fc0 sc0 ls0 ws0">emerged <span class="_ _d"></span><span class="ls1c">as <span class="ls1d">a <span class="_ _d"></span></span></span>popular <span class="_ _7"> </span>solution <span class="_ _d"></span>to <span class="_ _8"></span>the <span class="_ _8"></span><span class="ls1d">data <span class="_ _d"></span></span>glut <span class="_ _d"></span>problem </div><div class="t m12 x14 h14 y30 ff2 fs9 fc0 sc0 ls0 ws0">that <span class="_ _16"> </span>plagues <span class="_ _a"> </span>organizations <span class="_ _15"> </span>that <span class="_ _16"> </span>routinely <span class="_ _a"> </span>capture <span class="_ _9"> </span><span class="ls1e">but </span></div><div class="t m12 x31 h14 y31 ff2 fs9 fc0 sc0 ls0 ws0">have little <span class="_ _4"></span>time or inadequate <span class="_ _4"></span>tool <span class="_ _4"></span>to <span class="_ _4"></span>understand <span class="_ _4"></span>the <span class="ls1d">data </span></div><div class="t m12 x31 h14 y32 ff2 fs9 fc0 sc0 ls0 ws0">sufficiently. <span class="_ _14"> </span><span class="ls1c">The <span class="_ _4"></span></span>process <span class="_ _8"></span><span class="ls25">of <span class="_ _1e"> </span></span>KDD <span class="_ _8"></span>can <span class="_ _e"> </span>be <span class="_ _8"></span><span class="ls1e">viewed <span class="_ _7"> </span></span>as <span class="_ _1f"> </span><span class="ls1d">a </span></div><div class="t m12 x31 h14 y33 ff2 fs9 fc0 sc0 ls0 ws0">succession <span class="ls23">of <span class="_ _1e"> </span></span>stages, <span class="_ _d"></span>as <span class="_ _d"></span>depicted <span class="_ _4"></span><span class="ls1d">in <span class="_ _8"></span></span>Figure </div><div class="t m1a x32 h18 y34 ff2 fs8 fc0 sc0 ls26 ws0">1. </div><div class="t m12 x33 h14 y34 ff2 fs9 fc0 sc0 ls0 ws0">After <span class="_ _d"></span>the </div><div class="t m12 x31 h14 y35 ff2 fs9 fc0 sc0 ls1d ws0">data <span class="_ _d"></span><span class="ls0">has <span class="_ _8"></span><span class="ls1e">been <span class="_ _8"></span></span>prepared <span class="_ _7"></span></span>in <span class="_ _8"></span><span class="ls0">some <span class="_ _4"></span>appropriate <span class="_ _8"></span></span>file <span class="_ _d"></span><span class="ls0">format </span></div><div class="t m12 x31 h14 y36 ff2 fs9 fc0 sc0 ls0 ws0">and <span class="_"> </span>rendered <span class="_"> </span>consistent <span class="_ _a"> </span>(e.g., <span class="_"> </span>removal <span class="_ _a"> </span><span class="ls23">of <span class="_ _14"> </span><span class="ls1e">noisy <span class="_"> </span><span class="ls1d">data, </span></span></span></div><div class="t m12 x31 h14 y37 ff2 fs9 fc0 sc0 ls0 ws0">filling <span class="_ _16"> </span>in <span class="_ _16"> </span><span class="ls23">of <span class="_ _14"> </span></span>missing <span class="_ _16"> </span>values, <span class="_ _a"> </span>and <span class="_ _16"> </span>deleting <span class="_ _9"> </span>hopelessly </div><div class="t m12 x31 h14 y38 ff2 fs9 fc0 sc0 ls0 ws0">corrupted <span class="_ _a"> </span>records), <span class="_ _9"> </span>trends <span class="_ _9"> </span>and <span class="_ _9"> </span>patterns <span class="_ _15"> </span>can <span class="_ _9"> </span>then </div><div class="t m1b x34 h18 y39 ff2 fs8 fc0 sc0 ls15 ws0">be </div><div class="t m12 x31 h14 y3a ff2 fs9 fc0 sc0 ls0 ws0">generated, <span class="_"> </span>from <span class="_ _1f"> </span>which <span class="_"> </span>hypotheses <span class="_ _1f"> </span>can <span class="_"> </span>be <span class="_ _e"> </span>elicited <span class="_"> </span>for </div><div class="t m12 x31 h14 y3b ff2 fs9 fc0 sc0 ls0 ws0">interpretation <span class="_ _d"></span>and analysis. </div><div class="t m12 x35 h19 y3c ff7 fsd fc0 sc0 ls16 ws0">Figure </div><div class="t m1c x36 h1a y3c ff7 fs8 fc0 sc0 ls0 ws0">1 </div><div class="t m1d x37 h1b y3c ff3 fs11 fc0 sc0 ls0 ws0">: </div><div class="t m12 x38 h19 y3c ff7 fsd fc0 sc0 ls16 ws0">Knowledge <span class="_ _2"></span>Discovery <span class="_ _2"></span><span class="ls0">in <span class="_ _3"></span><span class="ls16">Databases </span></span></div><div class="t m12 x39 h19 y3d ff7 fsd fc0 sc0 ls16 ws0">Process </div><div class="t m12 x3a h14 y3e ff2 fs9 fc0 sc0 ls0 ws0">Research <span class="_ _4"></span><span class="ls1d">in <span class="_ _d"></span></span>KDD <span class="_ _4"></span>todate <span class="_ _4"></span>has <span class="_ _d"></span>focused largely <span class="_ _d"></span>on <span class="_ _d"></span>the </div><div class="t m12 x3b h14 y3f ff2 fs9 fc0 sc0 ls0 ws0">adaptation <span class="_ _1f"> </span><span class="ls23">of <span class="_ _9"> </span></span>machine <span class="_ _7"> </span><span class="ls1d">learning <span class="_ _7"> </span></span>techniques, <span class="_"> </span>with <span class="_ _1f"> </span>some </div><div class="t m12 x3b h14 y40 ff2 fs9 fc0 sc0 ls0 ws0">emphasis <span class="_ _19"> </span><span class="ls1d">on <span class="_ _20"> </span></span>statistical <span class="_ _19"> </span>analysis. <span class="_ _21"> </span><span class="ls1c">The <span class="_ _1b"> </span></span>use <span class="_ _22"> </span><span class="ls27">of </span></div><div class="t m12 x3b h14 y41 ff2 fs9 fc0 sc0 ls0 ws0">visualization <span class="_ _8"></span>technology <span class="_ _8"></span>has <span class="_ _7"> </span>largely <span class="_ _7"></span><span class="ls1e">been <span class="_ _1f"> </span></span>dismissed <span class="_ _7"> </span><span class="ls1e">as </span></div><div class="t m12 x3b h14 y42 ff2 fs9 fc0 sc0 ls0 ws0">being <span class="_ _d"></span>gimmicky <span class="_ _4"></span><span class="ls1d">in <span class="_ _8"></span></span>KDD <span class="_ _4"></span>work. </div><div class="t m12 x3c h14 y43 ff2 fs9 fc0 sc0 ls1d ws0">This <span class="_ _d"></span><span class="ls0">paper <span class="_ _7"></span>describes <span class="_ _4"></span></span>a </div><div class="t m12 x3b h14 y44 ff2 fs9 fc0 sc0 ls0 ws0">multi-dimensional <span class="_"> </span>visualization </div><div class="t m1e x3c h1c y45 ff2 fs11 fc0 sc0 ls0 ws0">(MDV) </div><div class="t m12 x3d h14 y45 ff2 fs9 fc0 sc0 ls0 ws0">technique <span class="_ _e"> </span>that </div><div class="t m12 x3b h14 y46 ff2 fs9 fc0 sc0 ls0 ws0">allows <span class="ls1d">a </span>user <span class="_ _4"></span><span class="ls1e">to <span class="_ _4"></span></span>visually <span class="ls1d">examine a </span>tabular <span class="_ _d"></span>database <span class="_ _4"></span>and </div><div class="t m12 x3b h14 y47 ff2 fs9 fc0 sc0 ls1e ws0">to <span class="_ _c"> </span><span class="ls0">formulate <span class="_ _c"> </span>query <span class="_ _14"> </span>interactively <span class="_ _14"> </span><span class="ls1d">and <span class="_ _c"> </span></span>visually. <span class="_ _23"> </span>In </span></div><div class="t m12 x3b h14 y48 ff2 fs9 fc0 sc0 ls0 ws0">particular, <span class="_ _b"> </span>the <span class="_ _c"> </span><span class="ls1e">MDV <span class="_ _18"> </span></span>technique <span class="_ _1c"> </span>has <span class="_ _18"> </span><span class="ls1e">been <span class="_ _18"> </span></span>used <span class="_ _c"> </span><span class="ls1d">in </span></div><div class="t m12 x3b h14 y49 ff2 fs9 fc0 sc0 ls0 ws0">exploratory <span class="_ _7"> </span><span class="ls1d">data <span class="_ _e"> </span></span>analysis <span class="_ _e"> </span>and <span class="_"> </span>KDD. <span class="_ _24"> </span><span class="ls1e">MDV <span class="_ _1e"> </span></span>has <span class="_ _a"> </span><span class="ls1e">been </span></div><div class="t m12 x3b h14 y4a ff2 fs9 fc0 sc0 ls0 ws0">realized <span class="_"> </span><span class="ls1d">in <span class="_ _a"> </span>a <span class="_ _1f"> </span></span>tool <span class="_ _a"> </span>named <span class="_"> </span>WinViz, <span class="_ _9"> </span>which <span class="_"> </span><span class="ls1d">a <span class="_ _a"> </span></span>user <span class="_"> </span><span class="ls1d">can </span></div><div class="t m12 x3b h14 y4b ff2 fs9 fc0 sc0 ls0 ws0">interactively <span class="_"> </span>explore <span class="_"> </span><span class="ls1d">a <span class="_"> </span></span>database <span class="_ _a"> </span>and <span class="_"> </span>identify <span class="_"> </span>patterns </div><div class="t m12 x3b h14 y4c ff2 fs9 fc0 sc0 ls0 ws0">and <span class="_ _7"></span>trends <span class="_ _1f"> </span>hidden <span class="_ _1f"> </span>in <span class="_ _1f"> </span>the <span class="_ _1f"> </span><span class="ls1d">data. <span class="_ _b"> </span><span class="ls1c">The <span class="_ _d"></span></span></span>user <span class="_ _7"> </span><span class="ls1d">can <span class="_ _1f"> </span></span>uncover </div><div class="t m12 x3b h14 y4d ff2 fs9 fc0 sc0 ls1e ws0">new <span class="ls0">properties </span></div><div class="t m1f x3e h1d y4e ff2 fs12 fc0 sc0 ls28 ws0">of </div><div class="t m12 x3f h14 y4e ff2 fs9 fc0 sc0 ls0 ws0">the <span class="ls1d">data <span class="_ _3"></span><span class="ls0">and detect any <span class="_ _4"></span>deviations. <span class="ls24">We </span></span></span></div><div class="t m12 x3b h14 y4f ff2 fs9 fc0 sc0 ls1e ws0">get <span class="_ _8"></span><span class="ls1d">a <span class="_ _d"></span><span class="ls0">picture <span class="_ _8"></span></span></span>of <span class="_ _8"></span><span class="ls1d">'what <span class="_ _4"></span>the <span class="_ _d"></span>data <span class="_ _d"></span>is <span class="_ _d"></span><span class="ls0">trying <span class="_ _8"></span></span></span>to <span class="_ _8"></span><span class="ls0">tell <span class="_ _8"></span></span>us' </div><div class="t m20 x40 h1e y50 ff3 fs13 fc0 sc0 ls0 ws0">- </div><div class="t m12 x41 h14 y50 ff2 fs9 fc0 sc0 ls0 ws0">the </div><div class="t m12 x3b h14 y51 ff2 fs9 fc0 sc0 ls0 ws0">observations <span class="_ _d"></span>can <span class="_ _8"></span>then <span class="_ _8"></span>be <span class="_ _d"></span>confirmed <span class="_ _8"></span>using <span class="_ _7"> </span>conventional </div><div class="t m12 x3b h14 y52 ff2 fs9 fc0 sc0 ls0 ws0">statistical analysis. </div><div class="t m21 x42 h1f y53 ff2 fs14 fc0 sc0 ls29 ws0">0730-3157/95 </div><div class="t m22 x1c h20 y53 ff2 fs15 fc0 sc0 ls0 ws0">$04.00 </div><div class="t m23 x43 h21 y53 ff5 fsf fc0 sc0 ls0 ws0">0 </div><div class="t m21 x44 h1f y53 ff2 fs14 fc0 sc0 ls29 ws0">1995 </div><div class="t m24 x45 h16 y53 ff2 fsf fc0 sc0 ls2a ws0">IEEE </div><div class="t m25 x46 h22 y54 ff1 fs16 fc0 sc0 ls2b ws0">26 </div><a class="l" rel='nofollow' onclick='return false;'><div class="d m26"></div></a></div><div class="pi" data-data='{"ctm":[1.563418,0.000000,0.000000,1.563418,0.000000,0.000000]}'></div></div> </body> </html>
评论
    相关推荐
    • 数据挖掘论文.rar
      数据挖掘领域前沿的42篇论文,我在iEEE数据库上找了好久才找齐的
    • 数据挖掘 论文 翻译 英汉
      这是一篇我自己找的一篇好的数据挖掘论文,并将其翻译成了汉语,也是我们的课堂作业。
    • 2017数据挖掘相关论文
      2017数据挖掘相关最新论文,包括数据清洗,实践分析等
    • 数据挖掘论文合集15
      Min-Max模糊神经网络的应用研究、ON-LINE REDUCING MACHINING ERRORS IN BORING OPERATIONBY FORECASTING COMPENSATORY CONTROL TECHNIQUE。。。
    • 模糊数据挖掘研究相关论文
      模糊数据挖掘是模糊集理论和数据挖掘技术的结合而成的一种新的挖掘方法,它解决了传统数据挖掘产生的“边界值过硬”的问题,提供了新的解决问题的方法。这里是几篇相关论文,希望对大家有所帮助!
    • 关于数据挖掘论文3
      网上搜集的关于数据挖掘论文,蛮全的,很大,所以分三个部分上传。这是第三个部分。
    • 数据挖掘论文精选
      内部精选十多个经典论文合成,包括数据挖掘、社团挖掘等
    • 数据挖掘经典论文.rar
      数据挖掘方向的五十篇经典论文数据挖掘学习必看论文
    • 数据挖掘论文.rar
      个人整理的有关机器学习和数据挖掘的IEEE,计算机学报上的论文,很全面的哟~
    • 国际期刊数据挖掘论文
      主要是关于SIG期刊上发表的关于kdd的论文,国际顶级会议发表的2015年的