ES \ub294 \uae30\ubcf8\uac12\uc73c\ub85c \uac80\uc0c9\ud0a4\uc6cc\ub4dc\uc758 \ube48\ub3c4\uc218, \ubb38\uc11c\uc0c1\uc758 \ud0a4\uc6cc\ub4dc\uc758 \ube48\ub3c4\uc218, \ubb38\uc11c\uc758 \ud06c\uae30 \ub4f1\uc744 \uae30\uc900\uc73c\ub85c score \ub97c \uc0b0\uc815\ud569\ub2c8\ub2e4.<\/p>\n \ubb38\uc11c \ub0b4\uc5d0\uc11c \uc790\uc8fc \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc5d0 \uac00\uc911\uce58\ub97c \ub192\uac8c \ubd80\uc5ec\ud569\ub2c8\ub2e4.<\/p>\n \ub9ce\uc740 \ubb38\uc11c\uc5d0\uc11c \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc758 \uac00\uc911\uce58\ub294 \ub0ae\ucd94\uace0, \uc77c\ubd80 \ubb38\uc11c\uc5d0\uc11c\ub9cc \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc5d0 \ub300\ud574 \ub192\uc740 \uac00\uc911\uce58\ub97c \ubd80\uc5ec\ud569\ub2c8\ub2e4.<\/p>\n \ubb38\uc11c\uc758 \uc885\ub958\ub97c \uac00\ub9ac\uc9c0 \uc54a\uace0 \uc790\uc8fc \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4\ub294 \ubcc4 \uc758\ubbf8\uc5c6\ub294 \ud754\ud55c \ub2e8\uc5b4\uc77c \uac00\ub2a5\uc131\uc774 \ub192\uae30 \ub54c\ubb38\uc785\ub2c8\ub2e4.<\/p>\n \ub450\uac1c \uc774\uc0c1\uc758 \ubb38\uc11c\uc5d0\uc11c \ub3d9\uc77c\ud55c \ud0a4\uc6cc\ub4dc\uac00 \ub3d9\uc77c\ud55c \ube48\ub3c4\uc218\ub85c \ub4f1\uc7a5\ud55c\ub2e4\uace0 \uac00\uc815\ud560 \ub54c, \ubb38\uc11c\uc758 \uae38\uc774\uac00 \uc791\uc740 \ubb38\uc11c\uc5d0 \ub354 \ub192\uc740 \uac00\uc911\uce58\ub97c \ubd80\uc5ec\ud569\ub2c8\ub2e4.<\/p>\n \ud558\uc9c0\ub9cc \uc1fc\ud551\ubab0\uc5d0\uc11c\ub294 \uac80\uc0c9\uc5b4\uc758 \ube48\ub3c4\uc218\ub9cc \uac00\uc9c0\uace0 score \ub97c \uc0b0\uc815\ud560 \uc218 \uc5c6\uc2b5\ub2c8\ub2e4.<\/p>\n \uac04\ub2e8\ud558\uac8c \uc544\ub798\uc5d0\uc11c \uc5ec\uae30<\/a> \uc5d0 \uc124\uba85\ub41c \uc11c\ubc84\uc124\uc815\uc744 \uae30\ucd08\ub85c \uc218\uc815\ud569\ub2c8\ub2e4.<\/p>\n \uc870\ud68c\ud0a4\uc6cc\ub4dc\uc640 \uad6c\ub9e4\ud0a4\uc6cc\ub4dc\ub97c \ub370\uc774\ud0c0\ub85c \uc785\ub825\ud574 \uc90d\ub2c8\ub2e4.<\/p>\n ElasticSearch score \ucee4\uc2a4\ud130\ub9c8\uc774\uc9d5 ES \ub294 \uae30\ubcf8\uac12\uc73c\ub85c BM25 \uc54c\uace0\ub9ac\uc998\uc744 \uc774\uc6a9\ud574 score \ub97c \uacc4\uc0b0\ud569\ub2c8\ub2e4. BM25 \uc54c\uace0\ub9ac\uc998 \uac80\uc0c9\ud0a4\uc6cc\ub4dc\uc758 \ube48\ub3c4\uc218, \ubb38\uc11c\uc0c1\uc758 \ud0a4\uc6cc\ub4dc\uc758 \ube48\ub3c4\uc218, \ubb38\uc11c\uc758 \ud06c\uae30 \ub4f1\uc744 \uae30\uc900\uc73c\ub85c score \ub97c \uc0b0\uc815\ud569\ub2c8\ub2e4. TF(Term Frequency) \ubb38\uc11c \ub0b4\uc5d0\uc11c \uc790\uc8fc \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc5d0 \uac00\uc911\uce58\ub97c \ub192\uac8c \ubd80\uc5ec\ud569\ub2c8\ub2e4. IDF(Inverse Document Frequency) \ub9ce\uc740 \ubb38\uc11c\uc5d0\uc11c \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc758 \uac00\uc911\uce58\ub294 \ub0ae\ucd94\uace0, \uc77c\ubd80 \ubb38\uc11c\uc5d0\uc11c\ub9cc \ub4f1\uc7a5\ud558\ub294 \ub2e8\uc5b4(Term) \uc5d0 \ub300\ud574 \ub192\uc740 \uac00\uc911\uce58\ub97c\u2026 Read More »<\/a><\/span><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[],"class_list":["post-962","post","type-post","status-publish","format-standard","hentry","category-elasticsearch"],"_links":{"self":[{"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/962","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=962"}],"version-history":[{"count":16,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/962\/revisions"}],"predecessor-version":[{"id":1005,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/962\/revisions\/1005"}],"wp:attachment":[{"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=962"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=962"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.skyer9.pe.kr\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=962"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}BM25<\/code> \uc54c\uace0\ub9ac\uc998\uc744 \uc774\uc6a9\ud574 score<\/code> \ub97c \uacc4\uc0b0\ud569\ub2c8\ub2e4.<\/p>\nBM25 \uc54c\uace0\ub9ac\uc998<\/h2>\n
TF(Term Frequency)<\/h3>\n
IDF(Inverse Document Frequency)<\/h3>\n
Field-length norm<\/h3>\n
score \ucee4\uc2a4\ud130\ub9c8\uc774\uc9d5<\/h2>\n
\uc544\uc774\ud3f0<\/code> \uc774\ub77c\ub294 \ud0a4\uc6cc\ub4dc\uc640 \uae30\uc874 \uace0\uac1d\ub4e4\uc774 \uac80\uc0c9 \ud6c4 \ud074\ub9ad\ud55c \uc218, \uad6c\ub9e4\ud55c \uc218\uac00 \ub192\uc740 \uc0c1\ud488\uc5d0 \uc6b0\uc120\uc21c\uc704\ub97c \ub354 \ubd80\uc5ec\ud560 \ud544\uc694\uac00 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\nfunction_score<\/code> \ub97c \uc774\uc6a9\ud574 \uc704 \uae30\ub2a5\uc744 \uad6c\ud604\ud574 \ubd05\ub2c8\ub2e4.<\/p>\n\ucc38\uc870 \uc0ac\uc774\ud2b8<\/h2>\n
\n
\uae30\ubcf8\uc124\uc815<\/h2>\n
\ud14c\uc774\ube14 \uc900\ube44<\/h2>\n
CREATE DATABASE db_test DEFAULT CHARACTER SET UTF8 COLLATE UTF8_GENERAL_CI;\n\nUSE db_test;\n\nDROP TABLE IF EXISTS items;\n\nCREATE TABLE items (\n itemid BIGINT(20) UNSIGNED NOT NULL,\n PRIMARY KEY (itemid),\n UNIQUE KEY unique_id (itemid),\n itemname VARCHAR(128) NOT NULL,\n viewKeywords VARCHAR(2048) NOT NULL,\n buyKeywords VARCHAR(2048) NOT NULL,\n lastupdate TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,\n regdate TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP\n);\n\nINSERT INTO items (itemid, itemname, viewKeywords, buyKeywords)\nVALUES (1, '\ube14\ub799 \uc544\uc774\ud3f0 \ucf00\uc774\uc2a4', '\uc544\uc774\ud3f0 \ucf00\uc774\uc2a4', '');\n\nINSERT INTO items (itemid, itemname, viewKeywords, buyKeywords)\nVALUES (2, '\ube14\ub799 \uac24\ub7ed\uc2dc \ucf00\uc774\uc2a4', '\uac24\ub7ed\uc2dc', '\uac24\ub7ed\uc2dc');\n\nINSERT INTO items (itemid, itemname, viewKeywords, buyKeywords)\nVALUES (3, '\ud551\ud06c \uc544\uc774\ud3f0 \ucf00\uc774\uc2a4', '\uc544\uc774\ud3f0', '');\n\nINSERT INTO items (itemid, itemname, viewKeywords, buyKeywords)\nVALUES (4, '\ud654\uc774\ud2b8 \uac24\ub7ed\uc2dc \ucf00\uc774\uc2a4', '', '');\n\nINSERT INTO items (itemid, itemname, viewKeywords, buyKeywords)\nVALUES (5, '\ub798\ub4dc \uc544\uc774\ud3f0 \ucf00\uc774\uc2a4', '\ucf00\uc774\uc2a4', '\uc544\uc774\ud3f0');<\/code><\/pre>\n\uc778\ub371\uc2a4 \uc900\ube44<\/h2>\n
curl -XDELETE http:\/\/localhost:9200\/items?pretty\n\ncurl -XPUT http:\/\/localhost:9200\/items?pretty -H 'Content-Type: application\/json' -d '{\n "settings" : {\n "index":{\n "analysis":{\n "analyzer":{\n "korean":{\n "type":"custom",\n "tokenizer":"seunjeon_default_tokenizer",\n "filter" : ["synonym", "stopword", "lowercase"]\n }\n },\n "filter" : {\n "synonym" : {\n "type" : "synonym",\n "synonyms_path" : "synonyms.txt"\n },\n "stopword" : {\n "type" : "stop",\n "stopwords_path" : "stopwords.txt"\n }\n },\n "tokenizer": {\n "seunjeon_default_tokenizer": {\n "index_eojeol": "true",\n "user_dict_path": "user_dict.csv",\n "index_poses": [\n "UNK", "EP", "I", "J", "M",\n "N", "SL", "SH", "SN", "VCP",\n "XP", "XS", "XR"\n ],\n "decompound": "true",\n "type": "seunjeon_tokenizer"\n }\n }\n }\n }\n },\n "mappings" : {\n "_doc" : {\n "properties" : {\n "itemid" : {\n "type" : "integer"\n },\n "itemname" : {\n "type" : "text",\n "analyzer": "korean"\n },\n "viewKeywords" : {\n "type" : "text",\n "analyzer": "korean"\n },\n "buyKeywords" : {\n "type" : "text",\n "analyzer": "korean"\n },\n "lastupdate" : {\n "type" : "date"\n },\n "regdate" : {\n "type" : "date"\n }\n }\n }\n }\n}'<\/code><\/pre>\n\uac00\uc911\uce58\ub97c \ubd80\uc5ec\ud55c \uac80\uc0c9<\/h2>\n
curl -X GET "localhost:9200\/items\/_search?pretty" -H 'Content-Type: application\/json' -d'{\n "query": {\n "function_score": {\n "query": { "match": { "itemname": "\uc544\uc774\ud3f0 \ucf00\uc774\uc2a4" } },\n "boost": "5",\n "functions": [\n {\n "filter": { "match": { "viewKeywords": "\uc544\uc774\ud3f0 \ucf00\uc774\uc2a4" } },\n "random_score": {},\n "weight": 23\n },\n {\n "filter": { "match": { "buyKeywords": "\uc544\uc774\ud3f0 \ucf00\uc774\uc2a4" } },\n "weight": 42\n }\n ],\n "max_boost": 42,\n "score_mode": "max",\n "boost_mode": "multiply",\n "min_score" : 0\n }\n }\n}'<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"