Data matrix
\nN record
\neach record has p features<\/p>\n
N is very large
\np is very large
\nboth of them
\nN no problem, p is very large
\np no peoblem, N is very large<\/p>\n
\u06a9\u0627\u0647\u0634 \u0648\u06cc\u0698\u06af\u06cc Feature Reduction
\n\u0641\u0631\u0636 \u06a9\u0646\u06cc\u0645 \u062f\u0631 \u0641\u0636\u0627\u06cc \u06f1\u06f0 \u0628\u0639\u062f\u06cc \u06a9\u0627\u0631 \u0645\u06cc \u06a9\u0646\u06cc\u0645 \u0645\u06cc \u062e\u0648\u0627\u0647\u06cc\u0645 \u0628\u0639\u062f \u0647\u0627 \u0631\u0627 \u06a9\u0627\u0647\u0634 \u062f\u0647\u06cc\u0645<\/p>\n
\u0628\u0631 \u0627\u0633\u0627\u0633 \u0645\u0633\u0627\u0644\u0647 \u06a9\u0627\u0631 \u0645\u06cc \u06a9\u0646\u0646\u062f<\/p>\n
supervised , unsupervised
\nminimum Information Loss<\/p>\n
\u0641\u0627\u0635\u0644\u0647 \u0628\u06cc\u0646 \u06a9\u0644\u0627\u0633 \u0647\u0627 \u0631\u0627 \u0647\u0645 \u0628\u0647\u062a\u0631 \u0627\u0633\u062a \u06a9\u0647 \u0628\u06cc\u0634\u062a\u0631 \u06a9\u0646\u06cc\u0645<\/p>\n
x1 \u062a\u0627 xn \u0645\u0634\u0627\u0647\u062f\u0627\u062a \u0645\u0627 \u0647\u0633\u062a\u0646\u062f
\n\u06a9\u0647 \u0641\u0636\u0627\u06cc p \u0628\u0639\u062f\u06cc \u0645\u0627 \u0647\u0633\u062a\u0646\u062f
\nG \u0645\u0627\u062a\u0631\u06cc\u0633 \u062a\u0628\u062f\u06cc\u0644 \u0647\u0633\u062a<\/p>\n
\u062f\u0627\u062f\u0647 \u0647\u0627\u06cc \u0628\u0627 \u062d\u062c\u0645 \u0632\u06cc\u0627\u062f Hign Dimentional \u0631\u0627 \u0628\u0627\u06cc\u062f \u062d\u062c\u0645\u0634 \u0631\u0627 \u06a9\u0645 \u06a9\u0646\u06cc\u0645<\/p>\n
\u062f\u0633\u062a\u0647 \u0628\u0646\u062f\u06cc<\/p>\n
Feature Selection : \u0627\u06af\u0631 p \u062a\u0627 \u0628\u0639\u062f \u062f\u0627\u0631\u0645 \u0641\u0642\u0637 \u0628\u0627 \u0641\u06cc\u0648\u0686\u0631 \u0647\u0627\u06cc \u0645\u0648\u062b\u0631 \u06a9\u0627\u0631 \u06a9\u0646\u06cc\u0645<\/p>\n
Visualization
\nData Compression
\nNoise Removal
\n—————-
\nApplication of feature reduction
\nFace recognition
\nHandwrittien digit recognition
\ntextmining
\nImage retrieval
\nMicroarray Data Analysis
\nProtein classification<\/p>\n
——————
\nFeature Reduction Algorithms :
\nUnsupervised :
\n-latent Semantinc Indexing ( LSI) : truncated SVD
\n– Independent Component Analysis (ICA)
\nPCA
\nCCA<\/p>\n
Supervised :
\nLDA<\/p>\n
Semi-supervised :
\nResearch Topics
\n——————–
\nLinear
\nLSI
\nPCA
\nLDA
\nCCA : cononical Correclation
\n—————————–<\/p>\n
PCA : principal Component Analysis<\/p>\n
<\/p>\n
http:\/\/www.sas.com\/data-visualization\/overview.html<\/a><\/p>\n <\/p>\n <\/p>\n","protected":false},"excerpt":{"rendered":" Data matrix N record each record has p features N is very large p is very large both of them N no problem, p is very large p no peoblem, N is very large \u06a9\u0627\u0647\u0634 \u0648\u06cc\u0698\u06af\u06cc Feature Reduction \u0641\u0631\u0636 \u06a9\u0646\u06cc\u0645 \u062f\u0631 \u0641\u0636\u0627\u06cc \u06f1\u06f0 \u0628\u0639\u062f\u06cc \u06a9\u0627\u0631 \u0645\u06cc \u06a9\u0646\u06cc\u0645 \u0645\u06cc \u062e\u0648\u0627\u0647\u06cc\u0645 \u0628\u0639\u062f \u0647\u0627 \u0631\u0627 \u06a9\u0627\u0647\u0634 \u062f\u0647\u06cc\u0645 \u0628\u0631 […]<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[21],"tags":[],"class_list":["post-553","post","type-post","status-publish","format-standard","hentry","category-clustering","category-21-id","post-seq-1","post-parity-odd","meta-position-corners","fix"],"_links":{"self":[{"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/posts\/553"}],"collection":[{"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=553"}],"version-history":[{"count":3,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/posts\/553\/revisions"}],"predecessor-version":[{"id":555,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=\/wp\/v2\/posts\/553\/revisions\/555"}],"wp:attachment":[{"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=553"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=553"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/vua.nadiran.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=553"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}