{"id":476450,"date":"2023-08-09T07:29:55","date_gmt":"2023-08-09T07:29:55","guid":{"rendered":""},"modified":"2023-09-05T11:12:45","modified_gmt":"2023-09-05T11:12:45","slug":"cosine-similarity","status":"publish","type":"wiki","link":"https:\/\/oneproxy.pro\/tr\/wiki\/cosine-similarity\/","title":{"rendered":"Kosin\u00fcs benzerli\u011fi"},"content":{"rendered":"<p>Kosin\u00fcs benzerli\u011fi, matematik ve do\u011fal dil i\u015flemede (NLP) bir i\u00e7 \u00e7arp\u0131m uzay\u0131ndaki s\u0131f\u0131r olmayan iki vekt\u00f6r aras\u0131ndaki benzerli\u011fi \u00f6l\u00e7en temel bir kavramd\u0131r. Bilgi eri\u015fimi, metin madencili\u011fi, \u00f6neri sistemleri ve daha fazlas\u0131 dahil olmak \u00fczere \u00e7e\u015fitli alanlarda yayg\u0131n olarak kullan\u0131lmaktad\u0131r. Bu makale Kosin\u00fcs benzerli\u011finin tarihini, i\u00e7 yap\u0131s\u0131n\u0131, t\u00fcrlerini, kullan\u0131mlar\u0131n\u0131 ve gelece\u011fe y\u00f6nelik perspektiflerini ele alacakt\u0131r.<\/p>\n<h2>Kosin\u00fcs benzerli\u011finin k\u00f6keninin tarihi ve ilk s\u00f6z\u00fc<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi kavram\u0131, \u0130svi\u00e7reli matematik\u00e7i Adrien-Marie Legendre&#039;nin eliptik integraller \u00fczerine yapt\u0131\u011f\u0131 \u00e7al\u0131\u015fman\u0131n bir par\u00e7as\u0131 olarak bunu tan\u0131tt\u0131\u011f\u0131 19. y\u00fczy\u0131l\u0131n ba\u015flar\u0131na kadar izlenebilmektedir. Daha sonra, 20. y\u00fczy\u0131lda Kosin\u00fcs benzerli\u011fi bilgi eri\u015fimi alan\u0131na girdi ve NLP, belgeleri ve metin benzerli\u011fini kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in yararl\u0131 bir \u00f6l\u00e7\u00fc olarak kullan\u0131ld\u0131.<\/p>\n<h2>Kosin\u00fcs benzerli\u011fi hakk\u0131nda detayl\u0131 bilgi. Konunun geni\u015fletilmesi Kosin\u00fcs benzerli\u011fi<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi, \u00e7ok boyutlu bir alanda kar\u015f\u0131la\u015ft\u0131r\u0131lan belgeleri veya metinleri temsil eden iki vekt\u00f6r aras\u0131ndaki a\u00e7\u0131n\u0131n kosin\u00fcs\u00fcn\u00fc hesaplar. A ve B gibi iki vekt\u00f6r aras\u0131ndaki Kosin\u00fcs benzerli\u011fini hesaplama form\u00fcl\u00fc \u015f\u00f6yledir:<\/p>\n<pre><div class=\"bg-black rounded-md mb-4\"><div class=\"flex items-center relative text-gray-200 bg-gray-800 px-4 py-2 text-xs font-sans justify-between rounded-t-md\"><span>css<\/span><button class=\"flex ml-auto gap-2\"><svg stroke=\"currentColor\" fill=\"none\" stroke-width=\"2\" viewbox=\"0 0 24 24\" stroke-linecap=\"round\" stroke-linejoin=\"round\" class=\"h-4 w-4\" height=\"1em\" width=\"1em\" ><path d=\"M16 4h2a2 2 0 0 1 2 2v14a2 2 0 0 1-2 2H6a2 2 0 0 1-2-2V6a2 2 0 0 1 2-2h2\"><\/path><rect x=\"8\" y=\"2\" width=\"8\" height=\"4\" rx=\"1\" ry=\"1\"><\/rect><\/svg>Kodu kopyala<\/button><\/div><div class=\"p-4 overflow-y-auto\"><code class=\"!whitespace-pre hljs language-css\" data-no-translation=\"\">Cosine Similarity(<span class=\"hljs-selector-tag\">A<\/span>, <span class=\"hljs-selector-tag\">B<\/span>) = (<span class=\"hljs-selector-tag\">A<\/span> \u00b7 <span class=\"hljs-selector-tag\">B<\/span>) \/ (||<span class=\"hljs-selector-tag\">A<\/span>|| * ||<span class=\"hljs-selector-tag\">B<\/span>||)\n<\/code><\/div><\/div><\/pre>\n<p>Neresi <code data-no-translation=\"\">(A \u00b7 B)<\/code> A ve B vekt\u00f6rlerinin nokta \u00e7arp\u0131m\u0131n\u0131 temsil eder ve <code data-no-translation=\"\">||A||<\/code> Ve <code data-no-translation=\"\">||B||<\/code> s\u0131ras\u0131yla A ve B vekt\u00f6rlerinin b\u00fcy\u00fckl\u00fckleridir (veya normlar\u0131d\u0131r).<\/p>\n<p>Kosin\u00fcs benzerli\u011fi -1 ila 1 aras\u0131nda de\u011fi\u015fir; -1 tam farkl\u0131l\u0131\u011f\u0131, 1 mutlak benzerli\u011fi ve 0 dikli\u011fi (benzerlik yok) belirtir.<\/p>\n<h2>Kosin\u00fcs benzerli\u011finin i\u00e7 yap\u0131s\u0131. Kosin\u00fcs benzerli\u011fi nas\u0131l \u00e7al\u0131\u015f\u0131r?<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi, metinsel verileri y\u00fcksek boyutlu bir uzayda say\u0131sal g\u00f6sterimlere (vekt\u00f6rlere) d\u00f6n\u00fc\u015ft\u00fcrerek \u00e7al\u0131\u015f\u0131r. Her boyut, veri k\u00fcmesindeki benzersiz bir terime kar\u015f\u0131l\u0131k gelir. Daha sonra iki belge aras\u0131ndaki benzerlik, kar\u015f\u0131l\u0131k gelen vekt\u00f6rler aras\u0131ndaki a\u00e7\u0131ya g\u00f6re belirlenir.<\/p>\n<p>Kosin\u00fcs benzerli\u011fini hesaplama s\u00fcreci a\u015fa\u011f\u0131daki ad\u0131mlar\u0131 i\u00e7erir:<\/p>\n<ol>\n<li>Metin \u00d6n \u0130\u015fleme: Durdurulan s\u00f6zc\u00fckleri, \u00f6zel karakterleri kald\u0131r\u0131n ve metni standartla\u015ft\u0131rmak i\u00e7in k\u00f6k ay\u0131rma veya lemmatizasyon ger\u00e7ekle\u015ftirin.<\/li>\n<li>Terim Frekans\u0131 (TF) Hesaplamas\u0131: Belgedeki her terimin s\u0131kl\u0131\u011f\u0131n\u0131 say\u0131n.<\/li>\n<li>Ters Belge S\u0131kl\u0131\u011f\u0131 (IDF) Hesaplamas\u0131: Nadir terimlere daha fazla a\u011f\u0131rl\u0131k vermek i\u00e7in t\u00fcm belgelerde her bir terimin \u00f6nemini \u00f6l\u00e7\u00fcn.<\/li>\n<li>TF-IDF Hesaplamas\u0131: Belgelerin nihai say\u0131sal temsilini elde etmek i\u00e7in TF ve IDF&#039;yi birle\u015ftirin.<\/li>\n<li>Kosin\u00fcs Benzerli\u011fi Hesaplamas\u0131: Belgelerin TF-IDF vekt\u00f6rlerini kullanarak Kosin\u00fcs benzerli\u011fini hesaplay\u0131n.<\/li>\n<\/ol>\n<h2>Kosin\u00fcs benzerli\u011finin temel \u00f6zelliklerinin analizi<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi, onu metin kar\u015f\u0131la\u015ft\u0131rma g\u00f6revleri i\u00e7in pop\u00fcler bir se\u00e7im haline getiren \u00e7e\u015fitli temel \u00f6zellikler sunar:<\/p>\n<ol>\n<li><strong>\u00d6l\u00e7ek De\u011fi\u015fmez<\/strong>: Kosin\u00fcs benzerli\u011fi vekt\u00f6rlerin b\u00fcy\u00fckl\u00fc\u011f\u00fcnden etkilenmez, bu da onu belge uzunluklar\u0131ndaki de\u011fi\u015fikliklere kar\u015f\u0131 dayan\u0131kl\u0131 k\u0131lar.<\/li>\n<li><strong>Yeterlik<\/strong>: Kosin\u00fcs benzerli\u011finin hesaplanmas\u0131, b\u00fcy\u00fck metin veri k\u00fcmeleri i\u00e7in bile hesaplama a\u00e7\u0131s\u0131ndan verimlidir.<\/li>\n<li><strong>Yorumlanabilirlik<\/strong>: Benzerlik puanlar\u0131 -1 ile 1 aras\u0131nda de\u011fi\u015fir ve sezgisel yorumlar sa\u011flar.<\/li>\n<li><strong>Metinsel Semantik Benzerlik<\/strong>: Kosin\u00fcs benzerli\u011fi, metinler aras\u0131ndaki anlamsal benzerli\u011fi dikkate alarak i\u00e7erik bazl\u0131 \u00f6nerilere ve k\u00fcmelemeye uygun hale getirir.<\/li>\n<\/ol>\n<h2>Kosin\u00fcs benzerli\u011fi t\u00fcrleri<\/h2>\n<p>Yayg\u0131n olarak kullan\u0131lan iki temel Kosin\u00fcs benzerli\u011fi t\u00fcr\u00fc vard\u0131r:<\/p>\n<ol>\n<li><strong>Klasik Kosin\u00fcs Benzerli\u011fi<\/strong>: Bu, belgelerin TF-IDF g\u00f6sterimi kullan\u0131larak daha \u00f6nce tart\u0131\u015f\u0131lan standart Kosin\u00fcs benzerli\u011fidir.<\/li>\n<li><strong>\u0130kili Kosin\u00fcs Benzerli\u011fi<\/strong>: Bu varyantta vekt\u00f6rler ikili olup, belgedeki terimlerin varl\u0131\u011f\u0131n\u0131 (1) veya yoklu\u011funu (0) g\u00f6sterir.<\/li>\n<\/ol>\n<p>\u0130\u015fte iki t\u00fcr\u00fcn kar\u015f\u0131la\u015ft\u0131rma tablosu:<\/p>\n<table>\n<thead>\n<tr>\n<th><\/th>\n<th>Klasik Kosin\u00fcs Benzerli\u011fi<\/th>\n<th>\u0130kili Kosin\u00fcs Benzerli\u011fi<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Vekt\u00f6r G\u00f6sterimi<\/td>\n<td>TF-IDF<\/td>\n<td>\u0130kili<\/td>\n<\/tr>\n<tr>\n<td>Yorumlanabilirlik<\/td>\n<td>Ger\u00e7ek de\u011ferli (-1&#039;den 1&#039;e)<\/td>\n<td>\u0130kili (0 veya 1)<\/td>\n<\/tr>\n<tr>\n<td>\u0130\u00e7in uygun<\/td>\n<td>Metin tabanl\u0131 uygulamalar<\/td>\n<td>Seyrek veri senaryolar\u0131<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>Kosin\u00fcs benzerli\u011fini kullanma yollar\u0131, kullan\u0131mla ilgili problemler ve \u00e7\u00f6z\u00fcmleri<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi \u00e7e\u015fitli alanlarda uygulamalar bulur:<\/p>\n<ol>\n<li><strong>Bilgi alma<\/strong>: Kosin\u00fcs benzerli\u011fi, dok\u00fcmanlar\u0131n bir sorguyla ilgisine g\u00f6re s\u0131ralanmas\u0131na yard\u0131mc\u0131 olarak arama motorlar\u0131n\u0131n verimli olmas\u0131n\u0131 sa\u011flar.<\/li>\n<li><strong>Belge K\u00fcmeleme<\/strong>: Daha iyi organizasyon ve analiz i\u00e7in benzer belgelerin bir arada grupland\u0131r\u0131lmas\u0131n\u0131 kolayla\u015ft\u0131r\u0131r.<\/li>\n<li><strong>\u0130\u015fbirlik\u00e7i Filtreleme<\/strong>: \u00d6neri sistemleri, benzer zevklere sahip kullan\u0131c\u0131lara \u00fcr\u00fcn \u00f6nermek i\u00e7in Kosin\u00fcs benzerli\u011fini kullan\u0131r.<\/li>\n<li><strong>\u0130ntihal Tespiti<\/strong>: Farkl\u0131 belgelerdeki benzer metin b\u00f6l\u00fcmlerini tan\u0131mlayabilir.<\/li>\n<\/ol>\n<p>Ancak Kosin\u00fcs benzerli\u011fi baz\u0131 durumlarda a\u015fa\u011f\u0131daki gibi zorluklarla kar\u015f\u0131la\u015fabilir:<\/p>\n<ul>\n<li><strong>K\u0131tl\u0131k<\/strong>: Y\u00fcksek boyutlu seyrek verilerle u\u011fra\u015f\u0131rken benzerlik puanlar\u0131 daha az bilgilendirici olabilir.<\/li>\n<li><strong>Dil Ba\u011f\u0131ml\u0131l\u0131\u011f\u0131<\/strong>: Kosin\u00fcs benzerli\u011fi, karma\u015f\u0131k dilbilgisi veya kelime d\u00fczenine sahip dillerde ba\u011flam\u0131 yakalayamayabilir.<\/li>\n<\/ul>\n<p>Bu sorunlar\u0131n \u00fcstesinden gelmek i\u00e7in, performans\u0131 art\u0131rmak amac\u0131yla boyut azaltma (\u00f6rne\u011fin, Tekil De\u011fer Ayr\u0131\u015ft\u0131rma kullan\u0131larak) ve s\u00f6zc\u00fck yerle\u015ftirme (\u00f6rne\u011fin, Word2Vec) gibi teknikler kullan\u0131l\u0131r.<\/p>\n<h2>Ana \u00f6zellikler ve benzer terimlerle di\u011fer kar\u015f\u0131la\u015ft\u0131rmalar<\/h2>\n<table>\n<thead>\n<tr>\n<th><\/th>\n<th>Kosin\u00fcs Benzerli\u011fi<\/th>\n<th>Jaccard Benzerli\u011fi<\/th>\n<th>\u00d6klid Mesafesi<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>\u00d6l\u00e7\u00fc Tipi<\/td>\n<td>Benzerlik<\/td>\n<td>Benzerlik<\/td>\n<td>Farkl\u0131l\u0131k<\/td>\n<\/tr>\n<tr>\n<td>Menzil<\/td>\n<td>-1&#039;e 1<\/td>\n<td>0&#039;dan 1&#039;e<\/td>\n<td>0&#039;dan \u221e&#039;a<\/td>\n<\/tr>\n<tr>\n<td>Uygulanabilirlik<\/td>\n<td>Metin kar\u015f\u0131la\u015ft\u0131rmas\u0131<\/td>\n<td>Kar\u015f\u0131la\u015ft\u0131rmay\u0131 ayarla<\/td>\n<td>Say\u0131sal vekt\u00f6rler<\/td>\n<\/tr>\n<tr>\n<td>Boyutluluk<\/td>\n<td>Y\u00fcksek boyutlu<\/td>\n<td>D\u00fc\u015f\u00fck boyutlu<\/td>\n<td>Y\u00fcksek boyutlu<\/td>\n<\/tr>\n<tr>\n<td>Hesaplama<\/td>\n<td>Verimli<\/td>\n<td>Verimli<\/td>\n<td>Hesaplama Yo\u011funlu\u011fu<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>Kosin\u00fcs benzerli\u011fine ili\u015fkin gelece\u011fin perspektifleri ve teknolojileri<\/h2>\n<p>Teknoloji ilerlemeye devam ettik\u00e7e Kosin\u00fcs benzerli\u011finin \u00e7e\u015fitli alanlarda de\u011ferli bir ara\u00e7 olarak kalmas\u0131 bekleniyor. Daha g\u00fc\u00e7l\u00fc donan\u0131m ve algoritmalar\u0131n ortaya \u00e7\u0131kmas\u0131yla birlikte Kosin\u00fcs benzerli\u011fi, b\u00fcy\u00fck veri k\u00fcmelerinin i\u015flenmesinde ve kesin \u00f6neriler sa\u011flanmas\u0131nda daha da verimli hale gelecektir. Ek olarak, do\u011fal dil i\u015fleme ve derin \u00f6\u011frenmede devam eden ara\u015ft\u0131rmalar, metin temsillerinin iyile\u015ftirilmesine yol a\u00e7arak benzerlik hesaplamalar\u0131n\u0131n do\u011frulu\u011funu daha da art\u0131rabilir.<\/p>\n<h2>Proxy sunucular\u0131 nas\u0131l kullan\u0131labilir veya Kosin\u00fcs benzerli\u011fiyle ili\u015fkilendirilebilir<\/h2>\n<p>OneProxy taraf\u0131ndan sa\u011flanan proxy sunucular\u0131, anonim ve g\u00fcvenli internet eri\u015fimini kolayla\u015ft\u0131rmada \u00e7ok \u00f6nemli bir rol oynar. Do\u011frudan Kosin\u00fcs benzerli\u011finden yararlanamasalar da, metin kar\u015f\u0131la\u015ft\u0131rmas\u0131 veya i\u00e7erik tabanl\u0131 filtreleme kullanan uygulamalarda yer alabilirler. \u00d6rne\u011fin, proxy sunucular, kullan\u0131c\u0131 tercihlerini kar\u015f\u0131la\u015ft\u0131rmak ve ilgili i\u00e7eri\u011fi \u00f6nermek i\u00e7in Kosin\u00fcs benzerli\u011finden yararlanarak \u00f6neri sistemlerinin performans\u0131n\u0131 art\u0131rabilir. Ayr\u0131ca, kullan\u0131c\u0131 sorgular\u0131 ve dizine eklenen belgeler aras\u0131ndaki benzerlik puanlar\u0131na dayal\u0131 olarak arama sonu\u00e7lar\u0131n\u0131 optimize ederek bilgi alma g\u00f6revlerine yard\u0131mc\u0131 olabilirler.<\/p>\n<h2>\u0130lgili Ba\u011flant\u0131lar<\/h2>\n<p>Kosin\u00fcs benzerli\u011fi hakk\u0131nda daha fazla bilgi i\u00e7in a\u015fa\u011f\u0131daki kaynaklara ba\u015fvurabilirsiniz:<\/p>\n<ol>\n<li><a href=\"https:\/\/en.wikipedia.org\/wiki\/Cosine_similarity\" target=\"_new\" rel=\"noopener nofollow\">Vikipedi \u2013 Kosin\u00fcs Benzerli\u011fi<\/a><\/li>\n<li><a href=\"https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.metrics.pairwise.cosine_similarity.html\" target=\"_new\" rel=\"noopener nofollow\">Scikit-learn \u2013 Kosin\u00fcs Benzerli\u011fi<\/a><\/li>\n<li><a href=\"https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.feature_extraction.text.TfidfVectorizer.html\" target=\"_new\" rel=\"noopener nofollow\">TfidfVectorizer \u2013 Sklearn Belgeleri<\/a><\/li>\n<li><a href=\"https:\/\/nlp.stanford.edu\/IR-book\/\" target=\"_new\" rel=\"noopener nofollow\">Bilgi Eri\u015fimine Giri\u015f \u2013 Manning, Raghavan, Sch\u00fctze<\/a><\/li>\n<\/ol>\n<p>Sonu\u00e7 olarak, Kosin\u00fcs benzerli\u011fi, NLP, bilgi eri\u015fimi ve \u00f6neri sistemlerinde geni\u015f bir uygulama yelpazesine sahip g\u00fc\u00e7l\u00fc bir matematiksel kavramd\u0131r. Basitli\u011fi, verimlili\u011fi ve yorumlanabilirli\u011fi onu \u00e7e\u015fitli metin tabanl\u0131 g\u00f6revler i\u00e7in pop\u00fcler bir se\u00e7im haline getiriyor ve teknolojide devam eden ilerlemelerin gelecekte yeteneklerini daha da geli\u015ftirmesi bekleniyor. \u0130\u015fletmeler ve ara\u015ft\u0131rmac\u0131lar Kosin\u00fcs benzerli\u011finin potansiyelinden yararlanmaya devam ettik\u00e7e OneProxy gibi proxy sunucular, g\u00fcvenli ve anonim internet eri\u015fimi sa\u011flarken bu uygulamalar\u0131 desteklemede hayati bir rol oynayacak.<\/p>","protected":false},"featured_media":468030,"menu_order":0,"template":"","meta":{"_acf_changed":false,"content-type":"","inline_featured_image":false,"footnotes":""},"class_list":["post-476450","wiki","type-wiki","status-publish","has-post-thumbnail","hentry"],"acf":{"faq_title":"Frequently Asked Questions about <mark>Cosine Similarity: A Comprehensive Guide<\/mark>","faq_items":[{"question":"What is Cosine similarity?","answer":"<p>Cosine similarity is a mathematical concept used to measure the similarity between two vectors in a multi-dimensional space. It is commonly applied in text analysis, recommendation systems, and information retrieval tasks.<\/p>"},{"question":"How does Cosine similarity work?","answer":"<p>Cosine similarity calculates the cosine of the angle between two vectors, representing the documents being compared. It ranges from -1 to 1, where -1 indicates complete dissimilarity, 1 indicates absolute similarity, and 0 indicates orthogonality (no similarity).<\/p>"},{"question":"What are the key features of Cosine similarity?","answer":"<p>Cosine similarity offers scale invariance, efficiency, interpretability, and the ability to measure textual semantic similarity.<\/p>"},{"question":"What types of Cosine similarity exist?","answer":"<p>There are two primary types: Classic Cosine Similarity, which uses TF-IDF representation, and Binary Cosine Similarity, which utilizes binary vectors.<\/p>"},{"question":"How can Cosine similarity be used?","answer":"<p>Cosine similarity finds applications in various fields, including information retrieval, document clustering, collaborative filtering, and plagiarism detection.<\/p>"},{"question":"What challenges does Cosine similarity face?","answer":"<p>Cosine similarity may encounter issues with sparsity and language dependence in certain scenarios. Techniques like dimensionality reduction and word embeddings can address these challenges.<\/p>"},{"question":"How does Cosine similarity compare to other similarity measures?","answer":"<p>Cosine similarity is distinct from Jaccard similarity and Euclidean distance in terms of range, applicability, dimensionality, and computation.<\/p>"},{"question":"What are the future perspectives of Cosine similarity?","answer":"<p>As technology advances, Cosine similarity is expected to remain a valuable tool with enhanced efficiency and accuracy in similarity calculations.<\/p>"},{"question":"How are proxy servers associated with Cosine similarity?","answer":"<p>While proxy servers like OneProxy don't directly utilize Cosine similarity, they can support applications that involve text comparison and content-based filtering, such as recommendation systems and information retrieval tasks. They also ensure secure internet access during these operations.<\/p>"}]},"_links":{"self":[{"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/wiki\/476450","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/wiki"}],"about":[{"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/types\/wiki"}],"version-history":[{"count":0,"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/wiki\/476450\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/media\/468030"}],"wp:attachment":[{"href":"https:\/\/oneproxy.pro\/tr\/wp-json\/wp\/v2\/media?parent=476450"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}