{"id":13884,"date":"2014-01-16T22:29:59","date_gmt":"2014-01-16T13:29:59","guid":{"rendered":"http:\/\/lab.synergy-marketing.co.jp\/?p=7520"},"modified":"2018-11-14T16:33:52","modified_gmt":"2018-11-14T07:33:52","slug":"solr-vs-elasticsearch-morelikethis","status":"publish","type":"post","link":"https:\/\/www.techscore.com\/blog\/2014\/01\/16\/solr-vs-elasticsearch-morelikethis\/","title":{"rendered":"Solr vs elasticsearch \u985e\u4f3c\u6587\u66f8\u691c\u7d22\uff08MoreLikeThis\uff09"},"content":{"rendered":"<p>\u3053\u3093\u306b\u3061\u306f\u3001\u99ac\u5834\u3067\u3059\u3002<br \/>\nSolr\u3068elasticsearch\u3092\u6bd4\u8f03\u3059\u308b\u30b7\u30ea\u30fc\u30ba\u3001\u6700\u7d42\u56de\u306f\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u6a5f\u80fd\u306b\u3064\u3044\u3066\u6bd4\u8f03\u3057\u307e\u3059\u3002<\/p>\n<p>Solr\u3068elasticsearch\u3001\u4e21\u65b9\u306e\u30d9\u30fc\u30b9\u3068\u306a\u3063\u3066\u3044\u308bLucene\u306b\u306fMoreLikeThis\u3068\u3044\u3046\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u306e\u6a5f\u80fd\u304c\u5b9f\u88c5\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u4e21\u8005\u3068\u3082\u5f53\u7136Lucene\u306eMoreLikeThis\u3092\u5229\u7528\u3057\u3066\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u6a5f\u80fd\u3092\u63d0\u4f9b\u3057\u3066\u3044\u307e\u3059\u304c\u3001API \u306e\u5f62\u5f0f\u306a\u3069\u306f\u304b\u306a\u308a\u7570\u306a\u308a\u307e\u3059\u3002<\/p>\n<h3> Solr\u306e\u5834\u5408 <\/h3>\n<p>Solr\u3067\u306f\u3001\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u306e\u30d7\u30ed\u30b0\u30e9\u30e0\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n<pre lang=\"scala\">\r\nimport org.apache.solr.client.solrj.impl.HttpSolrServer\r\nimport org.apache.solr.client.solrj.SolrQuery\r\nimport org.apache.solr.common.util.NamedList\r\nimport org.apache.solr.common.SolrDocument\r\n\r\n    object MoreLikeThis {\r\n        val url = \"http:\/\/localhost:8983\/solr\"\r\n        val count = 6\r\n\r\n        \/**\r\n         * id \u3067\u6307\u5b9a\u3055\u308c\u305f\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3068\u4f3c\u305f\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092count\u306e\u6570\u3060\u3051\u53d6\u5f97\u3059\u308b\r\n         * @param id\r\n         * @return \u4f3c\u305f\u8a18\u4e8b\u306eID\u306eList\r\n         *\/\r\n        def getRelatedDocs(id: Int, count: Int = count): List[String] = {\r\n\r\n          val server = new HttpSolrServer(url)\r\n\r\n          val query = new SolrQuery(\"id:\" + id)\r\n          query.set(\"mlt\", true)\r\n          query.set(\"mlt.fl\", \"title,body\")\r\n          query.set(\"mlt.midf\", 1)\r\n          query.set(\"mlt.mintf\", 1)\r\n          query.set(\"mlt.count\", count)\r\n          query.set(\"fl\", \"id,score,title,body\")\r\n\r\n          val response = server.query(query)\r\n          val doc = response.getResults.get(0)\r\n\r\n          response.getResponse().get(\"moreLikeThis\").asInstanceOf[NamedList[Object]]\r\n            .getVal(0).asInstanceOf[java.util.List[SolrDocument]]\r\n            .toList.map(_.get(\"id\").toString))\r\n\r\n        }\r\n\r\n      }\r\n<\/pre>\n<h3>elasticsearch\u306e\u5834\u5408<\/h3>\n<p>elasticsearch\u3067\u306f\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u306e\u30d7\u30ed\u30b0\u30e9\u30e0\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n<pre lang=\"scala\">\r\nimport scala.collection.JavaConversions._\r\nimport org.elasticsearch.client.transport.TransportClient\r\nimport org.elasticsearch.common.transport.InetSocketTransportAddress\r\nimport org.elasticsearch.common.settings.ImmutableSettings\r\n\r\nobject MoreLikeThis {\r\n  val count = 6\r\n\r\n    \/**\r\n   * id \u3067\u6307\u5b9a\u3055\u308c\u305f\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3068\u4f3c\u305f\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092count\u306e\u6570\u3060\u3051\u53d6\u5f97\u3059\u308b\r\n   * @param id\r\n   * @return \u4f3c\u305f\u8a18\u4e8b\u306eID\u306eList\r\n   *\/\r\n  def getRelatedDocs(id: String, count: Int = count): List[String] = {\r\n    val settings = ImmutableSettings.settingsBuilder()\r\n      .put(\"client.transport.sniff\", false).build()\r\n    val client = new TransportClient(settings)\r\n      .addTransportAddress(new InetSocketTransportAddress(\"localhost\", 9300))\r\n\r\n    val response=client.prepareMoreLikeThis(\"test\", \"docs\", id.toInt)\r\n      .setField(\"title\", \"body\")\r\n      .setSearchTypes(\"docs\").execute().actionGet()\r\n\r\n    response.getHits.getHits.take(count).map (_.id().toString).toList\r\n\r\n  }\r\n\r\n}\r\n<\/pre>\n<p>Solr\u3068\u540c\u3058\u3088\u3046\u306bmidf\/mintf \u306e\u8a2d\u5b9a\u30921\u306b\u3059\u308b\u3053\u3068\u306f\u3067\u304d\u308b\u306e\u3067\u3059\u304c\u3001\u305d\u306e\u5834\u5408\u7d50\u679c\u304c\u8fd4\u3063\u3066\u304d\u307e\u305b\u3093\u3002<\/p>\n<pre lang=\"scala\">\r\n\/\/ \u3053\u306e\u8a2d\u5b9a\u3060\u3068\u691c\u7d22\u7d50\u679c\u306f\uff10\u4ef6\u306b\u306a\u308b\r\nval response=client.prepareMoreLikeThis(\"test\", \"docs\", id.toInt)\r\n      .setMinDocFreq(1)\r\n      .setMinTermFreq(1)\r\n      .setField(\"title\", \"body\")\r\n      .setSearchTypes(\"docs\").execute().actionGet()\r\n\r\n<\/pre>\n<p>\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u306f\u3001\u6587\u66f8\u306eID\u3092\u6307\u5b9a\u3059\u308b\u65b9\u6cd5\u3060\u3051\u3067\u306f\u306a\u304f\u3001\u6587\u66f8\u306e\u30bf\u30a4\u30c8\u30eb\u3084\u672c\u6587\u305d\u306e\u3082\u306e\u3092\u6307\u5b9a\u3059\u308b\u65b9\u6cd5\u3082\u3042\u308a\u307e\u3059\u3002<\/p>\n<pre lang=\"scala\">\r\nimport scala.collection.JavaConversions._\r\nimport org.elasticsearch.client.transport.TransportClient\r\nimport org.elasticsearch.common.transport.InetSocketTransportAddress\r\nimport org.elasticsearch.common.settings.ImmutableSettings\r\nimport org.elasticsearch.index.query.{FilterBuilders,  QueryBuilders}\r\n\r\n\r\nobject MoreLikeThis {\r\n  val count = 6\r\n\r\n  def getRelatedDocs(id: String, title:String, body:String,  count: Int = count): List[String] = {\r\n    val settings = ImmutableSettings.settingsBuilder()\r\n      .put(\"client.transport.sniff\", false).build()\r\n    val client = new TransportClient(settings)\r\n      .addTransportAddress(new InetSocketTransportAddress(\"localhost\", 9300))\r\n\r\n    val query = QueryBuilders.moreLikeThisQuery(\"title\", \"body\")\r\n      .likeText(title + \"\\n\" + body)\r\n    val response = client.prepareSearch(\"test\").setQuery(query)\r\n      .setFilter(FilterBuilders.notFilter(FilterBuilders.idsFilter(\"docs\").addIds(id))) \/\/\u691c\u7d22\u7d50\u679c\u304b\u3089id\u306e\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u3092\u9664\u5916\u3059\u308b\r\n      .setFrom(0).setSize(count).setExplain(true).execute().actionGet\r\n\r\n    response.getHits.getHits.take(count).map (_.id().toString).toList\r\n\r\n  }\r\n\r\n}\r\n<\/pre>\n<p>\u4e0b\u8a18\u306e\u30d7\u30ed\u30b0\u30e9\u30e0\u306e\u7d50\u679c\u306e\u65b9\u304c\u3001Solr\u306e\u691c\u7d22\u7d50\u679c\u3068\u4f3c\u3066\u3044\u307e\u3059\uff08\u304c\u3001\u5b8c\u5168\u306b\u4e00\u81f4\u3057\u307e\u305b\u3093...)\u3002\u5185\u90e8\u7684\u306b\u4f55\u304b\u9055\u3046\u306e\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002<\/p>\n<h3>\u307e\u3068\u3081<\/h3>\n<p>\u691c\u7d22API \u306f\u3001Solr \u304c\u3000Map \u3092\u69cb\u7bc9\u3057\u3066\u6e21\u3059\u306e\u306b\u5bfe\u3057\u3066\u3001elasticsearch\u306f\u30e1\u30bd\u30c3\u30c9\u3092\u3064\u306a\u3052\u3066\u884c\u304f\u5f62\u306a\u306e\u3067\u66f8\u3044\u3066\u3044\u3066\u5fc3\u5730\u3088\u3044\u3067\u3059\u3002\u305f\u3060\u3001\u3053\u3053\u3067\u3082elasticsearch \u306e\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u306eID\u3092\u6307\u5b9a\u3057\u3066\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u3092\u3059\u308b\u65b9\u6cd5\u306f\u30de\u30cb\u30e5\u30a2\u30eb\u306b\u306f\u66f8\u3044\u3066\u306a\u304f\u3001API\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u304b\u3089\u300c\u767a\u898b\u300d\u3057\u305f\u3082\u306e\u3067\u3059\u306e\u3067\u3001elasticsearch\u306f\u307e\u3060\u307e\u3060\u767a\u5c55\u4e2d\u3060\u3068\u611f\u3058\u307e\u3057\u305f\u3002<\/p>\n<p>\uff13\u56de\u306b\u308f\u305f\u3063\u3066\u3001Solr \u3068\u3000elasticsearch\u306e\u6bd4\u8f03\u3092\u884c\u3044\u307e\u3057\u305f\u3002\u5b9f\u969b\u306b\u3069\u3061\u3089\u3092\u63a1\u7528\u3059\u308b\u304b\u6c7a\u5b9a\u3059\u308b\u5834\u5408\u3001\u904b\u7528\u3084\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u3082\u8981\u56e0\u3068\u306a\u308b\u3068\u601d\u3044\u307e\u3059\u304c\u3001\u3053\u306e\u8a18\u4e8b\u304c\u3059\u3053\u3057\u3067\u3082\u307f\u306a\u3055\u3093\u306e\u53c2\u8003\u306b\u306a\u308c\u3070\u3068\u601d\u3044\u307e\u3059\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u3053\u3093\u306b\u3061\u306f\u3001\u99ac\u5834\u3067\u3059\u3002<br \/>\nSolr\u3068elasticsearch\u3092\u6bd4\u8f03\u3059\u308b\u30b7\u30ea\u30fc\u30ba\u3001\u6700\u7d42\u56de\u306f\u985e\u4f3c\u6587\u66f8\u691c\u7d22\u6a5f\u80fd\u306b\u3064\u3044\u3066\u6bd4\u8f03\u3057\u307e\u3059\u3002<br \/><a href=\"https:\/\/www.techscore.com\/blog\/2014\/01\/16\/solr-vs-elasticsearch-morelikethis\/\">\u7d9a\u304d\u3092\u8aad\u3080...<\/a><\/p>\n","protected":false},"author":23,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[255,18],"tags":[],"_links":{"self":[{"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/posts\/13884"}],"collection":[{"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/users\/23"}],"replies":[{"embeddable":true,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/comments?post=13884"}],"version-history":[{"count":4,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/posts\/13884\/revisions"}],"predecessor-version":[{"id":13984,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/posts\/13884\/revisions\/13984"}],"wp:attachment":[{"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/media?parent=13884"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/categories?post=13884"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.techscore.com\/blog\/wp-json\/wp\/v2\/tags?post=13884"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}