{"id":464,"date":"2018-09-17T12:00:00","date_gmt":"2018-09-17T03:00:00","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/example-of-c-parameter-of-svm-in-r\/"},"modified":"2024-10-12T22:35:32","modified_gmt":"2024-10-12T13:35:32","slug":"example-of-c-parameter-of-svm-in-r","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/example-of-c-parameter-of-svm-in-r\/","title":{"rendered":"R \u3067 SVM \u306e C \u30d1\u30e9\u30e1\u30fc\u30bf\u306b\u3064\u3044\u3066\u5177\u4f53\u4f8b\u3092\u793a\u3059"},"content":{"rendered":"\n

SVM\uff08\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\uff09\u306e\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf C \u306b\u3064\u3044\u3066\u3002<\/p>\n\n\n\n\n\n\n\n

SVM \u306e C \u3068\u306f\uff1f<\/h2>\n\n\n\n

SVM\uff08\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\uff09\u306e\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf C \u3068\u306f\u4f55\u304b\uff1f<\/p>\n\n\n\n

\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf C \u306f\u8aa4\u5206\u985e\u3092\u8a31\u5bb9\u3059\u308b\u6307\u6a19\u3002<\/p>\n\n\n\n

C \u304c\u5c0f\u3055\u3044\u3068\u8aa4\u5206\u985e\u3092\u8a31\u5bb9\u3059\u308b\u3002<\/p>\n\n\n\n

\u5927\u304d\u3044\u3068\u8aa4\u5206\u985e\u3092\u8a31\u5bb9\u3057\u306a\u3044\u3002<\/p>\n\n\n\n

C \u304c\u5927\u304d\u3044\u5834\u5408\u306f\u3001\u8907\u96d1\u306b\u5206\u985e\u3057\u3066\u3044\u308b\u3002<\/p>\n\n\n\n

\u8907\u96d1\u306b\u5206\u985e\u3059\u308b\u3068\u3001\u5b66\u7fd2\u3057\u3066\u3044\u308b\u30c7\u30fc\u30bf\u3067\u306f\u8aa4\u5206\u985e\u305b\u305a\u304d\u308c\u3044\u306b\u5206\u985e\u3067\u304d\u308b\u3002<\/p>\n\n\n\n

\u3057\u304b\u3057\u3001\u6c4e\u7528\u6027\u306f\u4e0b\u304c\u308b\u3002<\/p>\n\n\n\n

\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3067\u306f\u5206\u985e\u6027\u80fd\u304c\u4e0b\u304c\u308b\u3002<\/p>\n\n\n\n

C \u3092\u5c0f\u3055\u304f\u3057\u3066\u3001\u8aa4\u308a\u3092\u3042\u308b\u7a0b\u5ea6\u8a31\u5bb9\u3059\u308b\u3068\u6c4e\u5316\u6027\u80fd\u304c\u4e0a\u304c\u308b\u3002<\/p>\n\n\n\n

\"\"\/<\/a><\/figure>\n\n\n\n

<\/p>\n\n\n\n

\u51fa\u5178\uff1aTokyo.R 41 \u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\u3067\u773c\u93e1\u3063\u5a18\u5206\u985e\u30b7\u30b9\u30c6\u30e0\u69cb\u7bc9<\/a><\/p>\n\n\n\n

SVM \u306e C \u4f8b\u793a\u306e\u305f\u3081\u306e\u30b5\u30f3\u30d7\u30eb\u30c7\u30fc\u30bf\u6e96\u5099<\/h2>\n\n\n\n

ISLR\u30d1\u30c3\u30b1\u30fc\u30b8\u306eOJ\u30c7\u30fc\u30bf\u3092\u4f7f\u3046\u3002<\/p>\n\n\n\n

OJ\u30c7\u30fc\u30bf\u306f\u30b7\u30c8\u30e9\u30b9\u30d2\u30eb\uff08CH\uff09\u3068\u30df\u30cb\u30c3\u30c4\u30e1\u30a4\u30c9\uff08MM\uff09\u3069\u3061\u3089\u306e\u30aa\u30ec\u30f3\u30b8\u30b8\u30e5\u30fc\u30b9\u3092\u8cb7\u3063\u305f\u304b\u306e\u30c7\u30fc\u30bf\u3002<\/p>\n\n\n\n

\u3069\u3061\u3089\u306e\u30aa\u30ec\u30f3\u30b8\u30b8\u30e5\u30fc\u30b9\u8cb7\u3046\u304b\u3092\u4e88\u6e2c\u3059\u308b\u5206\u985e\u5668\u3092\u4f5c\u6210\u3059\u308b\u3068\u3044\u3046\u304a\u984c\u3002<\/p>\n\n\n\n

\u6700\u521d\u4e00\u56de\u3060\u3051\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u3002<\/p>\n\n\n\n

install.packages<\/span>(<\/span>\"ISLR\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u547c\u3073\u51fa\u3059\u306e\u304clibrary()\u3060\u3002<\/p>\n\n\n\n
library<\/span>(<\/span>ISLR)<\/span>\n<\/code><\/pre>\n\n\n\n
\u5b66\u7fd2\u30bb\u30c3\u30c8\u306e\u305f\u3081\u306b1070\u4ef6\u306e\u30c7\u30fc\u30bf\u306e3\u5206\u306e2\u3092\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3059\u308b\u3002<\/p>\n\n\n\n
\u540c\u3058\u7d50\u679c\u304c\u5f97\u3089\u308c\u308b\u305f\u3081\u306bseed\u3092\u6c7a\u3081\u3066\u304a\u304f\u3002<\/p>\n\n\n\n
set.seed<\/span>(<\/span>20180916<\/span>)<\/span>\nsub <-<\/span> c<\/span>(<\/span>sample<\/span>(<\/span>1<\/span>:<\/span>1070<\/span>,<\/span> round<\/span>(<\/span>1070<\/span>*<\/span>2<\/span>\/<\/span>3<\/span>)))<\/span>\n<\/code><\/pre>\n\n\n\n
\uff1e\uff1e\u3082\u3046\u7d71\u8a08\u3067\u60a9\u3080\u306e\u306f\u7d42\u308f\u308a\u306b\u3057\u307e\u305b\u3093\u304b\uff1f\u00a0<\/a><\/strong><\/span><\/p>\r\n\"\"<\/a>\r\n
\u21911\u4e07\u4eba\u4ee5\u4e0a\u306e\u533b\u7642\u5f93\u4e8b\u8005\u304c\u8cfc\u8aad\u4e2d<\/span><\/strong><\/span><\/p><\/div>
SVM \u306e C \u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u5909\u3048\u3066\u7d50\u679c\u3092\u78ba\u8a8d\u3057\u3066\u307f\u308b<\/h2>\n\n\n\n
\u30b3\u30b9\u30c8\u306e\u30c7\u30d5\u30a9\u30eb\u30c8\u306f 1 \u3067\u3001\u6307\u5b9a\u3057\u306a\u3044\u3068 1 \u3067\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
model1 <-<\/span> svm<\/span>(<\/span>Purchase ~<\/span> .,<\/span> data =<\/span> OJ[<\/span>sub,],<\/span> cross=<\/span>10<\/span>)<\/span>\n(<\/span>svm.confusion.tr1 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model1),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr1))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr1)<\/span>\npred1 <-<\/span> predict<\/span>(<\/span>model1,<\/span> newdata=<\/span>OJ[<\/span>-<\/span>sub,])<\/span>\n(<\/span>svm.confusion1 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred1))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion1))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion1)<\/span>\n<\/code><\/pre>\n\n\n\n
\u5b66\u7fd2\u30bb\u30c3\u30c8\u306e\u6b63\u7b54\u7387\u306f86.0\uff05\u3067\u3001\u30c6\u30b9\u30c8\u30bb\u30c3\u30c8\u3067\u306f81.5%\u3060\u3002<\/p>\n\n\n\n
><\/span> (<\/span>svm.confusion.tr1 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model1),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nCH  MM\nCH 393<\/span>  65<\/span>\nMM  35<\/span> 220<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr1))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr1)<\/span>\n[<\/span>1<\/span>]<\/span> 0.8597475<\/span>\n><\/span> (<\/span>svm.confusion1 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred1))<\/span>\npred1\nCH  MM\nCH 200<\/span>  25<\/span>\nMM  41<\/span>  91<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion1))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion1)<\/span>\n[<\/span>1<\/span>]<\/span> 0.8151261<\/span>\n<\/code><\/pre>\n\n\n\n
\u30b3\u30b9\u30c8\u309210\u500d\u306e10\u306b\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
model2 <-<\/span> svm<\/span>(<\/span>Purchase ~<\/span> .,<\/span> data =<\/span> OJ[<\/span>sub,],<\/span> cross=<\/span>10<\/span>,<\/span> cost=<\/span>10<\/span>)<\/span>\n(<\/span>svm.confusion.tr2 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model2),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr2))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr2)<\/span>\npred2 <-<\/span> predict<\/span>(<\/span>model2,<\/span> newdata=<\/span>OJ[<\/span>-<\/span>sub,])<\/span>\n(<\/span>svm.confusion2 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred2))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion2))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion2)<\/span>\n<\/code><\/pre>\n\n\n\n
\u5b66\u7fd2\u30bb\u30c3\u30c8\u306e\u6b63\u7b54\u7387\u306f87.0%\u307e\u3067\u4e0a\u304c\u308b\u304c\u3001\u30c6\u30b9\u30c8\u30bb\u30c3\u30c8\u3067\u306f78.4%\u306b\u4e0b\u304c\u3063\u305f\u3002<\/p>\n\n\n\n
\u30b3\u30b9\u30c8\u3092\u9ad8\u304f\u3059\u308b\u3068\u6c4e\u5316\u6027\u80fd\u306f\u4e0b\u304c\u308b\u3053\u3068\u304c\u308f\u304b\u308b\u3002<\/p>\n\n\n\n
><\/span> (<\/span>svm.confusion.tr2 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model2),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nCH  MM\nCH 394<\/span>  59<\/span>\nMM  34<\/span> 226<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr2))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr2)<\/span>\n[<\/span>1<\/span>]<\/span> 0.8695652<\/span>\n><\/span> (<\/span>svm.confusion2 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred2))<\/span>\npred2\nCH  MM\nCH 193<\/span>  32<\/span>\nMM  45<\/span>  87<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion2))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion2)<\/span>\n[<\/span>1<\/span>]<\/span> 0.7843137<\/span>\n<\/code><\/pre>\n\n\n\n
\u30b3\u30b9\u30c8\u30920.1\u306b\u3057\u3066\u307f\u308b\u3068\u3069\u3046\u304b\uff1f<\/p>\n\n\n\n
model3 <-<\/span> svm<\/span>(<\/span>Purchase ~<\/span> .,<\/span> data =<\/span> OJ[<\/span>sub,],<\/span> cross=<\/span>10<\/span>,<\/span> cost=<\/span>0.1<\/span>)<\/span>\n(<\/span>svm.confusion.tr3 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model3),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr3))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr3)<\/span>\npred3 <-<\/span> predict<\/span>(<\/span>model3,<\/span> newdata=<\/span>OJ[<\/span>-<\/span>sub,])<\/span>\n(<\/span>svm.confusion3 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred3))<\/span>\nsum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion3))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion3)<\/span>\n<\/code><\/pre>\n\n\n\n
\u30b3\u30b9\u30c81\u306b\u6bd4\u3079\u308b\u3068\u3001\u5b66\u7fd2\u30bb\u30c3\u30c8\u3067\u3082\u30c6\u30b9\u30c8\u30bb\u30c3\u30c8\u3067\u3082\u6b63\u7b54\u7387\u306f\u4e0b\u304c\u308b\u3002<\/p>\n\n\n\n
\u30b3\u30b9\u30c8\u3092\u4e0b\u3052\u904e\u304e\u308b\u3068\u6027\u80fd\u304c\u60aa\u304f\u306a\u308b\u3053\u3068\u304c\u308f\u304b\u308b\u3002<\/p>\n\n\n\n
><\/span> (<\/span>svm.confusion.tr3 <-<\/span> table<\/span>(<\/span>fitted<\/span>(<\/span>model3),<\/span>OJ[<\/span>sub,]$<\/span>Purchase))<\/span>\nCH  MM\nCH 383<\/span>  80<\/span>\nMM  45<\/span> 205<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion.tr3))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion.tr3)<\/span>\n[<\/span>1<\/span>]<\/span> 0.8246844<\/span>\n><\/span> (<\/span>svm.confusion3 <-<\/span> table<\/span>(<\/span>OJ[<\/span>-<\/span>sub,]$<\/span>Purchase,<\/span> pred3))<\/span>\npred3\nCH  MM\nCH 201<\/span>  24<\/span>\nMM  46<\/span>  86<\/span>\n><\/span> sum<\/span>(<\/span>diag<\/span>(<\/span>svm.confusion3))<\/span>\/<\/span>sum<\/span>(<\/span>svm.confusion3)<\/span>\n[<\/span>1<\/span>]<\/span> 0.8039216<\/span>\n<\/code><\/pre>\n\n\n\n
\u307e\u3068\u3081<\/h2>\n\n\n\n
\u4ee5\u4e0a\u3001\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\u306e\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf C \u3092\u5909\u3048\u3066\u307f\u3066\u3001\u5206\u985e\u5668\u306e\u6027\u80fd\u3092\u78ba\u8a8d\u3057\u3066\u307f\u305f\u3002<\/p>\n\n\n\n
\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf\u306f\u4e0a\u3052\u904e\u304e\u3066\u3082\u4e0b\u3052\u904e\u304e\u3066\u3082\u3088\u304f\u306a\u3044\u3053\u3068\u304c\u308f\u304b\u3063\u305f\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
SVM\uff08\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30fc\u30de\u30b7\u30f3\uff09\u306e\u30b3\u30b9\u30c8\u30d1\u30e9\u30e1\u30fc\u30bf C \u306b\u3064\u3044\u3066\u3002<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"swell_btn_cv_data":"","_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[5,103],"tags":[],"class_list":["post-464","post","type-post","status-publish","format-standard","hentry","category-r","category-103"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/464","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/comments?post=464"}],"version-history":[{"count":2,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/464\/revisions"}],"predecessor-version":[{"id":2737,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/464\/revisions\/2737"}],"wp:attachment":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/media?parent=464"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/categories?post=464"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/tags?post=464"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}