{"id":8727,"date":"2017-03-17T15:00:14","date_gmt":"2017-03-17T06:00:14","guid":{"rendered":"http:\/\/sitest.jp\/blog\/?p=8727"},"modified":"2017-03-17T15:05:44","modified_gmt":"2017-03-17T06:05:44","slug":"apache-mahout-%e3%82%92%e4%bd%bf%e3%81%a3%e3%81%a6%e7%b0%a1%e5%8d%98%e3%81%aa%e3%83%ac%e3%82%b3%e3%83%a1%e3%83%b3%e3%83%89%e5%87%a6%e7%90%86%e3%82%92%e8%a9%a6%e3%81%97%e3%81%a6%e3%81%bf%e3%81%be","status":"publish","type":"post","link":"https:\/\/sitest.jp\/blog\/?p=8727","title":{"rendered":"Apache Mahout \u3092\u4f7f\u3063\u3066\u7c21\u5358\u306a\u30ec\u30b3\u30e1\u30f3\u30c9\u51e6\u7406\u3092\u8a66\u3057\u3066\u307f\u305f"},"content":{"rendered":"<p>\u30ec\u30b3\u30e1\u30f3\u30c9\u30a8\u30f3\u30b8\u30f3\u306f\u3001Amazon\u306e\u30aa\u30b9\u30b9\u30e1\u5546\u54c1\u6a5f\u80fd\u304c\u6709\u540d\u3067\u3059\u3002<br \/>\nAmazon\u304cWeb\u696d\u754c\u3067\u3053\u308c\u3060\u3051\u306e\u5f71\u97ff\u529b\u304c\u3042\u308b\u7406\u7531\u3068\u3057\u3066\u306f\u3001\u30ec\u30b3\u30e1\u30f3\u30c9\u30a8\u30f3\u30b8\u30f3\u3092\u6700\u521d\u306b\u30d3\u30b8\u30cd\u30b9\u3068\u3057\u3066\u5c55\u958b\u3055\u305b\u305f\u3053\u3068\u306e\u8981\u56e0\u304c\u5927\u304d\u3044\u3068\u601d\u308f\u308c\u307e\u3059\u3002<br \/>\n\u305d\u3053\u3067\u3001\u4eca\u56de\u306fApache Mahout \u3092\u4f7f\u3063\u3066\u7c21\u5358\u306a\u30ec\u30b3\u30e1\u30f3\u30c9\u51e6\u7406\u3092\u8a66\u3057\u3066\u307f\u307e\u3057\u305f\u3002<\/p>\n<p><!--more--><\/p>\n<div class=\"frame_headding\">\n<ul>\n<li>\n<h3 style=\"text-align: left;\">\u25a0 \u5b9f\u884c\u3057\u3066\u307f\u308b<\/h3>\n<\/li>\n<\/ul>\n<\/div>\n<p>Apache Mahout \u306b\u3088\u308b\u7c21\u5358\u306a\u30ec\u30b3\u30e1\u30f3\u30c9\u51e6\u7406\u306f\u3001\u5165\u529b\u30c7\u30fc\u30bf\u3092\u30b3\u30f3\u30d0\u30fc\u30c8\u3057\u305f DataModel \u3092\u30ec\u30b3\u30e1\u30f3\u30c9\u51e6\u7406\u3092\u30b3\u30f3\u30d0\u30fc\u30c8\u3057\u305f Recommender \u306b\u8a2d\u5b9a\u3057\u3066\u3001recommend \u30e1\u30bd\u30c3\u30c9\u306b\u30ec\u30b3\u30e1\u30f3\u30c9\u5bfe\u8c61\u306e\u30e6\u30fc\u30b6\u30fcID\u3068\u5bfe\u8c61\u3068\u306a\u308b\u30ec\u30b3\u30e1\u30f3\u30c9\u30a2\u30a4\u30c6\u30e0\u6570\u3092\u6307\u5b9a\u3057\u307e\u3059\u3002<\/p>\n<p>\u4eca\u56de\u306fCSV \u30d5\u30a1\u30a4\u30eb\u3092\u5165\u529b\u30c7\u30fc\u30bf\u3068\u3059\u308bFileDataModel\u3068GenericItemBasedRecommender\u3092\u4f7f\u7528\u3057\u3066\u307f\u307e\u3059\u3002<\/p>\n<p>\u5165\u529b CSV \u30d5\u30a1\u30a4\u30eb\u30d5\u30a9\u30fc\u30de\u30c3\u30c8\u306f\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n<pre><code>\u30e6\u30fc\u30b6\u30fcID, \u30a2\u30a4\u30c6\u30e0ID, \u8a55\u4fa1\u306e\u5024\n<\/code><\/pre>\n<p>\nGenericItemBasedRecommender \u306f\u30b3\u30f3\u30b9\u30c8\u30e9\u30af\u30bf\u5185\u3067\u6307\u5b9a\u3059\u308b Similarity \u306b\u3088\u3063\u3066\u985e\u4f3c\u6027\u306e\u8a08\u7b97\u65b9\u6cd5\u3092\u8a2d\u5b9a\u3067\u304d\u307e\u3059\u3002<br \/>\nMahout \u306b\u306f\u8907\u6570\u306e Similarity \u304c\u7528\u610f\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n<pre><code>EuclideanDistanceSimilarity : \u30e6\u30fc\u30af\u30ea\u30c3\u30c9\u8ddd\u96e2\nUncenteredCosineSimilarity : \u30b3\u30b5\u30a4\u30f3\u985e\u4f3c\u5ea6\nCityBlockSimilarity : \u30de\u30f3\u30cf\u30c3\u30bf\u30f3\u8ddd\u96e2\nPearsonCorrelationSimilarity : \u30d4\u30a2\u30bd\u30f3\u76f8\u95a2\n<\/code><\/pre>\n<p>\n\u306a\u3069\u304c\u6319\u3052\u3089\u308c\u307e\u3059\u3002<\/p>\n<p>\u8a00\u8a9e\u89e3\u6790\u306a\u3069\u306f\u3001\u4f5c\u6210\u3057\u305f\u30c7\u30fc\u30bf\u306e\u884c\u5217\u304c\u758e\u540d\u884c\u5217\u306b\u306a\u308a\u60c5\u5831\u3092\u5bc6\u306b\u3059\u308b\u305f\u3081SVD\uff08\u7279\u7570\u5024\u5206\u89e3\uff09\u306a\u3069\u306e\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u304c\u4f7f\u7528\u3055\u308c\u307e\u3059\u3002<\/p>\n<p>\u307e\u305f\u3001\u5354\u8abf\u30d5\u30a3\u30eb\u30bf\u30ea\u30f3\u30b0\u3067\u306fSlope One \u7b49\u306e\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u304c\u4f7f\u7528\u3055\u308c\u307e\u3059\u3002<br \/>\n\u3053\u308c\u3089\u306eSVD\u3084Slope One \u7b49\u306e\u672c\u683c\u7684\u306a\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3092\u4f7f\u7528\u3059\u308b\u306b\u306f\u3001Recommender\uff08SVDRecommender \u3084 SlopeOneRecommender\uff09\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002<\/p>\n<p>\u305d\u308c\u3067\u306f\u3001\u5358\u7d14\u306a\u4f8b\u3092\u5b9f\u884c\u3057\u3066\u307f\u307e\u3059\u3002<\/p>\n<p>\u5b9f\u884c\u4f8b \uff08\u30e6\u30fc\u30b6\u30fcID=1 \u306e\u30ec\u30b3\u30e1\u30f3\u30c9\u30a2\u30a4\u30c6\u30e0\u3092\u51fa\u529b\uff09<br \/>\n\u30b3\u30de\u30f3\u30c9\u3092\u53e9\u304f\u3068\u4ee5\u4e0b\u306e\u7d50\u679c\u304c\u51fa\u3066\u304d\u307e\u3057\u305f\u3002<\/p>\n<pre><code>result : 4, 3.6665667\n<\/code><\/pre>\n<p>\n\u30a2\u30a4\u30c6\u30e0ID=4 \u304c\u30ec\u30b3\u30e1\u30f3\u30c9\u3055\u308c\u305f\u30a2\u30a4\u30c6\u30e0\u3067\u3059\u3002<\/p>\n<p>\u5165\u529b\u30d5\u30a1\u30a4\u30eb\u306e\u30c7\u30fc\u30bf\u304c\u5358\u7d14\u3059\u304e\u305f\u305f\u3081<br \/>\n1\u30a2\u30a4\u30c6\u30e0\u3057\u304b\u51fa\u529b\u3055\u308c\u307e\u305b\u3093\u3067\u3057\u305f\u3002<\/p>\n<p>\u5165\u529b\u30d5\u30a1\u30a4\u30eb\u4f8b<\/p>\n<pre><code>1,2,5\n2,2,5\n1,3,5\n1,5,2\n2,1,4\n2,4,3\n2,5,1\n1,1,4\n\u30fb\u30fb\u30fb\n<\/code><\/pre>\n<p><\/p>\n<p>\u5927\u898f\u6a21\u306e\u5165\u529b\u30c7\u30fc\u30bf\u3092\u4f7f\u3046\u3068 \u985e\u4f3c\u6027 \u306b\u3088\u308b\u9055\u3044\u304c\u51fa\u3066\u304d\u307e\u3059\u3002<\/p>\n<p>\u5b9f\u884c\u4f8b2\uff08\u30e6\u30fc\u30af\u30ea\u30c3\u30c9\u8ddd\u96e2 EuclideanDistanceSimilarity\uff09<\/p>\n<pre><code>\u60c5\u5831: Processed 55131 user\nresult : 149, 2.2220574\nresult : 143, 2.2179555\nresult : 133, 2.2135546\nresult : 137, 2.2050555\nresult : 145, 2.205735\n<\/code><\/pre>\n<p>\n\u5b9f\u884c\u4f8b3\uff08\u30b3\u30b5\u30a4\u30f3\u985e\u4f3c\u5ea6 UncenteredCosineSimilarity\uff09<\/p>\n<pre><code>\u60c5\u5831: Processed 55131 users\nresult : 114, 4.388584\nresult : 117, 4.3185726\nresult : 129, 4.2214856\nresult : 138, 4.219598\nresult : 148, 4.108547\n<\/code><\/pre>\n<p>\n\u5b9f\u884c\u4f8b4\uff08\u30de\u30f3\u30cf\u30c3\u30bf\u30f3\u8ddd\u96e2 CityBlockSimilarity\uff09<\/p>\n<pre><code>\u60c5\u5831: Processed 55131 users\nresult : 117, 3.3547886\nresult : 129, 3.259705\nresult : 148, 3.1355408\nresult : 114, 3.1283574\nresult : 150, 3.1052058\n<\/code><\/pre>\n<p><!--more--><\/p>\n<div class=\"frame_headding\">\n<h3 style=\"text-align: left;\">\u25a0 \u6700\u5f8c\u306b<\/h3>\n<\/div>\n<p>\u3053\u306e\u3088\u3046\u306bApache Mahout\u3092\u5229\u7528\u3059\u308c\u3070\u3001\u30ec\u30b3\u30e1\u30f3\u30c9\u30a8\u30f3\u30b8\u30f3\u306a\u3069\u306e\u8907\u96d1\u306a\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3092\u7d44\u3080\u5fc5\u8981\u304c\u306a\u304f\u3001\u5927\u898f\u6a21\u306a\u30c7\u30fc\u30bf\u3082Hadoop\u3084Spark\u306e\u74b0\u5883\u4e0b\u306b\u3066\u52d5\u304b\u305b\u3070\u3001\u7c21\u5358\u306b\u5b9f\u73fe\u3067\u304d\u307e\u3059\u3002\u4e00\u5ea6\u8a66\u3057\u3066\u307f\u3066\u306f\u5982\u4f55\u3067\u3057\u3087\u3046\u304b\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u30ec\u30b3\u30e1\u30f3\u30c9\u30a8\u30f3\u30b8\u30f3\u306f\u3001Amazon\u306e\u30aa\u30b9\u30b9\u30e1\u5546\u54c1\u6a5f\u80fd\u304c\u6709\u540d\u3067\u3059\u3002 Amazon\u304cWeb\u696d\u754c\u3067\u3053\u308c\u3060\u3051\u306e\u5f71\u97ff\u529b\u304c\u3042\u308b\u7406\u7531\u3068\u3057\u3066\u306f\u3001\u30ec\u30b3\u30e1\u30f3\u30c9\u30a8\u30f3\u30b8\u30f3\u3092\u6700\u521d\u306b\u30d3\u30b8\u30cd\u30b9\u3068\u3057\u3066\u5c55\u958b\u3055\u305b\u305f\u3053\u3068\u306e\u8981\u56e0\u304c\u5927\u304d\u3044\u3068\u601d\u308f\u308c\u307e\u3059\u3002 \u305d\u3053\u3067\u3001\u4eca\u56de\u306fApache Mahout \u3092\u4f7f\u3063\u3066\u7c21\u5358\u306a\u30ec\u30b3\u30e1\u30f3\u30c9\u51e6\u7406\u3092\u8a66\u3057\u3066\u307f\u307e\u3057\u305f\u3002<\/p>\n","protected":false},"author":1,"featured_media":8740,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[25],"tags":[],"acf":false,"_links":{"self":[{"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/posts\/8727"}],"collection":[{"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/comments?post=8727"}],"version-history":[{"count":25,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/posts\/8727\/revisions"}],"predecessor-version":[{"id":9141,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/posts\/8727\/revisions\/9141"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/media\/8740"}],"wp:attachment":[{"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/media?parent=8727"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/categories?post=8727"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sitest.jp\/blog\/wp-json\/wp\/v2\/tags?post=8727"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}