{"id":14185,"date":"2019-03-29T07:15:38","date_gmt":"2019-03-29T05:15:38","guid":{"rendered":"https:\/\/www.dase-analytics.com\/blog\/?p=14185"},"modified":"2019-04-04T12:04:07","modified_gmt":"2019-04-04T10:04:07","slug":"5-sposobov-na-vizualizovanie-dat-pomocou-r-programming","status":"publish","type":"post","link":"https:\/\/www.dase-analytics.com\/blog\/sk\/5-sposobov-na-vizualizovanie-dat-pomocou-r-programming\/","title":{"rendered":"5 sp\u00f4sobov na vizualizovanie d\u00e1t pomocou R programming"},"content":{"rendered":"

V poslednom \u010dl\u00e1nku som p\u00edsal o sp\u00f4sobe, ako z\u00edska\u0165 d\u00e1ta z Google Analytics pomocou R. V dne\u0161nom sa pozrieme na mo\u017enosti vizu\u00e1liz\u00e1cie d\u00e1t. Budeme pou\u017e\u00edva\u0165 kni\u017enicu ggplot2 a predstav\u00edme si 5 zauj\u00edmav\u00fdch tipov, respekt\u00edve grafov, ktor\u00e9 v\u00e1m m\u00f4\u017eu v\u00fdrazne pom\u00f4c\u0165.<\/b><\/p>\n

Pred\u00fdm ako za\u010dneme som chcel V\u00e1m e\u0161te uk\u00e1za\u0165 nieko\u013eko zauji\u00edmav\u00fdch zdrojov na va\u0161e vzdel\u00e1vanie a pomoc pri pr\u00e1ci z R-kom ke\u010f\u017ee bol o to z\u00e1ujem a men\u0161ie technick\u00e9 probl\u00e9my s na\u0161im komentovac\u00edm syst\u00e9mom mi znemo\u017enili to zazdiela\u0165.<\/p>\n

V\u00fdborn\u00fdm zdrojom na pr\u00e1cu R a Google Analytics je str\u00e1nka www.dartistics.com<\/a>. Je to str\u00e1nka \u0161peci\u00e1lne pre nad\u0161encov a profesion\u00e1lov z digit\u00e1lnej analytiky a je to skvel\u00fd \u0161tartovac\u00ed bod aby ste z\u00edskali va\u010d\u0161iu istotu pri pr\u00e1ci z R. n\u00e1sledne doporu\u010dujem pri pr\u00e1ci z R pou\u017e\u00edva\u0165 \u0161tandardne str\u00e1nku stackoverflow.com \u00a0<\/a>kde si viete dohlada\u0165 odpovede na va\u0161e ot\u00e1zky. Som presved\u010den\u00fd, \u017ee to je najlep\u0161ia cesta ako sa nau\u010di\u0165 pracova\u0165 s R lebo mus\u00edte si spravi\u0165 tu n\u00e1mahu pr\u00eds\u0165 na veci, ktor\u00e9 chcete zisti\u0165 a potom si mus\u00edte e\u0161te poradi\u0165 aj s t\u00fdm ako to spravi\u0165. Ale v\u00fdsledok bude skvel\u00fd pre v\u00e1\u0161 v\u00fdvoj a pr\u00e1cu s R.<\/p>\n

A toto je \u010fal\u0161\u00ed zauj\u00edmav\u00fd zdroj na pr\u00e1cu s R a vizualiz\u00e1ciou –\u00a0https:\/\/www.statmethods.net<\/a>\u00a0.<\/p>\n

Tak a teraz po\u010fme priamo k na\u0161ej t\u00e9me. Aby sme mohli d\u00e1ta vizualizova\u0165, <\/span>najsk\u00f4r ich potrebujeme ma\u0165 dostupn\u00e9<\/b>. Op\u00e4\u0165 budeme pracova\u0165 s <\/span>d\u00e1tami z Google Analytics<\/b>.<\/span><\/p>\n

Ni\u017e\u0161ie vid\u00edte \u0161trukt\u00faru n\u00e1\u0161ho query. Ak ste ne\u010d\u00edtali predch\u00e1dzaj\u00faci \u010dl\u00e1nok o tom, ako vytvori\u0165 query na z\u00edskanie d\u00e1t z Google Analytics, m\u00f4\u017eete si ho pre\u010d\u00edta\u0165 tu.<\/a><\/span><\/p>\n

Na\u0161e query je nasledovn\u00e9:<\/span><\/p>\n

gadata <- google_analytics_4(view_id, <\/span><\/p>\n

\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0date_range = c(„2018-09-01“, „2019-03-15“),<\/span><\/p>\n

\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0metrics = c(„sessions“,“pageviewsPerSession“), \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0<\/span><\/p>\n

\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0dimensions = c(„date“, „deviceCategory“),<\/span><\/p>\n

\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0max = -1)<\/span><\/p>\n

Tabu\u013eka by mala vyzera\u0165 nasledovne:<\/span><\/p>\n

\"\"<\/p>\n

Osobne sa mi nep\u00e1\u010di, \u017ee <\/span>pageviewsPerSession s\u00fa s viacer\u00fdmi desatinn\u00fdmi \u010d\u00edslami<\/b>, tak si ich e\u0161te predt\u00fdm uprav\u00edme na cel\u00e9 \u010d\u00edslo.<\/span><\/p>\n

To sprav\u00edme nasleduj\u00facim sp\u00f4sobom:<\/span><\/p>\n

gadata$pageviewsPerSession <- \u00a0round(gadata$pageviewsPerSession, digits = 0)<\/span><\/p>\n

Parameter digits vlastne ur\u010duje na ko\u013eko desatinn\u00fdch \u010d\u00edsel \u00a0chceme zaokr\u00fahli\u0165 .<\/span><\/p>\n

NOTE:<\/b><\/p>\n

Ak sa chcete pohra\u0165 s in\u00fdmi metrikami alebo dimenziami, sta\u010d\u00ed ich vlo\u017ei\u0165 do vytvoren\u00e9ho query a z\u00edska\u0165 d\u00e1ta, s ktor\u00fdmi chcete pracova\u0165. Tieto som vybral len ako uk\u00e1\u017eku na pr\u00e1cu s ggplot2, aby ste videli, ako t\u00fato kni\u017enicu pou\u017ei\u0165 na vizualiz\u00e1ciu d\u00e1t a reprodukovanie.<\/b><\/p>\n

Ak\u00e9 grafy si uk\u00e1\u017eeme?<\/span><\/h2>\n
    \n
  1. Boxplot<\/span><\/li>\n
  2. Stack bar Chart<\/span><\/li>\n
  3. Time Series<\/span><\/li>\n
  4. Scatterplot<\/span><\/li>\n
  5. Histogram<\/span><\/li>\n<\/ol>\n

    Tak a po\u010fme si ich postupne predstavi\u0165 . <\/span><\/p>\n

    1.<\/strong>Vizualiz\u00e1cia pomocou boxplot grafu<\/h3>\n

    Boxplot<\/b> je <\/span>grafick\u00e9 zn\u00e1zornenie \u0161tatistick\u00fdch \u00fadajov na z\u00e1klade minima, prv\u00e9ho kvartilu, medi\u00e1nu, tretieho kvartilu a maxima<\/b>. Term\u00edn <\/span>\u201e<\/span>krabicov\u00fd graf“ poch\u00e1dza zo skuto\u010dnosti, \u017ee graf vyzer\u00e1 ako obd\u013a\u017enik s \u010diarami siahaj\u00facimi zhora nadol. <\/span><\/p>\n

    Tento \u0161t\u00fdl grafu r\u00e1d vyu\u017e\u00edvam na pozorovanie zmien napr\u00edklad v po\u010dte rel\u00e1cii, ktor\u00e9 ved\u00fa na web. Viete si r\u00fdchlo v\u0161imn\u00fa\u0165, \u017ee priemer\/medi\u00e1n je po\u010das niektor\u00fdch obdob\u00ed v\u00e4\u010d\u0161\u00ed. \u00a0M\u00f4\u017eete tie\u017e spozorova\u0165 nejak\u00e9 anom\u00e1lie, ktor\u00e9 s\u00fa v boxplote zv\u00fdraznen\u00e9 bodkami. <\/span><\/p>\n

    V ka\u017edom pr\u00edpade v\u00e1m to d\u00e1va insight, \u017ee by bolo dobr\u00e9 dan\u00e9 v\u00fdkyvy alebo n\u00e1rasty analyzova\u0165. Viete tak napr\u00edklad <\/span>ozrejmi\u0165 v\u00fdkon marketingovej kampane<\/b>, ke\u010f zist\u00edte, \u017ee k v\u00e1m prich\u00e1dza neakt\u00edvna n\u00e1v\u0161teva, ktorou m\u00f4\u017eu by\u0165 napr\u00edklad boti.<\/span><\/p>\n

    \"\"<\/p>\n

    Zdroj obr\u00e1zku: IPA Slovakia<\/span><\/p>\n

    A teraz si po\u010fme vytvori\u0165 n\u00e1\u0161 graf: \u00a0<\/span><\/p>\n

    boxplot <- ggplot(gadata, aes(x=deviceCategory, y=sessions, fill=deviceCategory)) + geom_boxplot()<\/span><\/p>\n

    \"\"<\/p>\n

    Ka\u017ed\u00e1 vizualiz\u00e1cia za\u010d\u00edna funkciou ggplot() a vn\u00fatri danej funkcie je na prvom mieste \u010fal\u0161ia tabulka, ktor\u00fa sme si nazvali gadata. Potom tam je \u010das s n\u00e1zvom \u00a0aes(), presnej\u0161ie aesthetics. <\/span><\/p>\n

    V \u010dasti aes() vlastne \u0161pecifikujeme, \u010do, kde a ako chceme ma\u0165. V na\u0161om pr\u00edpade si chceme vizualizova\u0165 zariadenia na ose x a rel\u00e1cie na ose y. <\/span>Parameter fill= n\u00e1m zas rozsegmentuje boxplot pod\u013ea zariaden\u00ed.<\/b> Geom_boxplot u\u017e len ur\u010duje sp\u00f4sob, ak\u00fdm grafom chceme dan\u00e9 d\u00e1ta vizualizova\u0165.<\/span><\/p>\n

    A cel\u00e9 to vyzer\u00e1 asi takto:<\/span><\/p>\n

    # Calculate the group means with aggregate<\/span><\/p>\n

    means <- aggregate(sessions ~ deviceCategory, gadata, mean )<\/span><\/p>\n

    plotchart <- gg + geom_text(data = means, aes(label = sessions, y = sessions + 0.20))<\/span><\/p>\n

    2.<\/strong>Vizualiz\u00e1cia pomocou stack chart grafu<\/h3>\n

    Stohovan\u00fd graf <\/b>(alebo stohovan\u00fd st\u013apcov\u00fd graf) je graf, ktor\u00fd pou\u017e\u00edva st\u013apce na porovnanie kateg\u00f3ri\u00ed \u00fadajov so schopnos\u0165ou rozobra\u0165 a porovna\u0165 \u010dasti celku. Ka\u017ed\u00fd st\u013apec v grafe predstavuje celok. Segmenty v st\u013apci predstavuj\u00fa r\u00f4zne \u010dasti alebo kateg\u00f3rie tohto celku.<\/span><\/p>\n

     <\/p>\n

    Ide o skvel\u00fd graf napr\u00edklad na porovnanie rel\u00e1ci\u00ed \u010di transakci\u00ed z kampan\u00ed alebo \u010dohoko\u013evek, \u010do v\u00e1m pom\u00f4\u017ee zodpoveda\u0165 va\u0161u ot\u00e1zku a z\u00edska\u0165 dan\u00fa inform\u00e1ciu. <\/span><\/p>\n

    \"\"<\/p>\n

    Na obr\u00e1zku vy\u0161\u0161ie <\/span>vizualizujeme pomer n\u00e1v\u0161tev z desktopu, mobilu a tabletu<\/b>.Pou\u017eili sme aj trendov\u00fa l\u00edniu, ktor\u00e1 n\u00e1m uk\u00e1\u017ee, \u017ee mobil za\u010dal by\u0165 zauj\u00edmavej\u0161\u00edm zariaden\u00edm, cez ktor\u00fd prich\u00e1dzaj\u00fa \u013eudia na n\u00e1\u0161 web. Je d\u00f4le\u017eit\u00e9 analyzova\u0165, pre\u010do tomu tak je a \u010di je v\u00e1\u0161 web pripraven\u00fd na tak\u00fdto traffic z mobilu z poh\u013eadu u\u017e\u00edvatelnosti. Ak m\u00e1te ecommerce str\u00e1nku, bolo by zauj\u00edmav\u00e9 rovnak\u00fdm sp\u00f4sobom vizualizova\u0165 pomer n\u00e1kupov z mobilu, desktopu a tabletu. <\/span><\/p>\n

    Uk\u00e1\u017eme si ako tento graf vytvori\u0165:<\/span><\/p>\n

    stackbar <- ggplot(data=gadata, aes(x=date, y=sessions, fill=deviceCategory)) +<\/span><\/p>\n

    \u00a0geom_bar(stat=“identity“) + geom_smooth(aes(date, sessions))<\/span><\/p>\n

    plot(stackbar)<\/span><\/p>\n

    Podobne, ako pri boxplote, pou\u017e\u00edvame funkciu ggplot() a n\u00e1sledne definujeme d\u00e1ta, ktor\u00e9 chceme vizualizova\u0165. Prv\u00e1 \u010das\u0165 je \u00faplne toto\u017en\u00e1 \u00a0s na\u0161\u00edm prv\u00fdm grafom. Tentokr\u00e1t ale nebudeme vizualizova\u0165 pomoc geom_boxplot, ale geom_bar. T\u00fdm p\u00e1dom z\u00edskame graf vy\u0161\u0161ie. <\/span><\/p>\n

    Na vizualiz\u00e1ciu trendu som tie\u017e pou\u017eil geom_smooth. Na konci \u010dl\u00e1nku sa s vami podel\u00edm o str\u00e1nky, kde sa viete nau\u010di\u0165, ako vytvori\u0165 graf, ktor\u00fd potrebujete.<\/span><\/p>\n

    PS: Ak by ste chceli vizualizova\u0165 transakcie podobn\u00fd sp\u00f4sobom, sta\u010d\u00ed len namiesto rel\u00e1ci\u00ed vlo\u017ei\u0165 d\u00e1ta z transakci\u00ed. <\/span><\/p>\n

    3.<\/strong>Vizualiz\u00e1cia pomocou time series grafu<\/h3>\n

    \u010eal\u0161\u00edm zauj\u00edmav\u00fdm grafom je<\/span> time series graf<\/b>. <\/span>Graf \u010dasov\u00fdch radov <\/span>zobrazuje d\u00e1tov\u00e9 body v postupn\u00fdch \u010dasov\u00fdch intervaloch<\/b>. Ka\u017ed\u00fd bod na grafe zodpoved\u00e1 \u010dasu a meran\u00e9mu mno\u017estvu.<\/span><\/p>\n

    V\u0161eobecne plat\u00ed, \u017ee horizont\u00e1lna os grafu sa pou\u017e\u00edva na vykreslenie \u010dasu a <\/span>vertik\u00e1lne osi ur\u010duj\u00fa hodnoty premennej<\/b> (v na\u0161om pr\u00edpade metriky, ako transakcie\/rel\u00e1cie\/po\u010det str\u00e1nok za rel\u00e1ciu), ktor\u00e1 sa meria. <\/span><\/p>\n

    ggplot(gadata, aes(date)) + <\/span><\/p>\n

    \u00a0geom_line(aes(y = sessions, colour = „sessions“)) +<\/span><\/p>\n

    \u00a0geom_line(aes(y = pageviewsPerSession, colour = „pageviewsPerSession“))<\/span><\/p>\n

    \"\"<\/p>\n

    4. Vizualiz\u00e1cia pomocou scatter plotu<\/h3>\n

    Rozptylov\u00fd graf <\/b>je mno\u017eina bodov vynesen\u00fdch na horizont\u00e1lnej a vertik\u00e1lnej osi.<\/span><\/p>\n

    Rozptylov\u00e9 grafy s\u00fa d\u00f4le\u017eit\u00e9 v \u0161tatistike, preto\u017ee <\/span>m\u00f4\u017eu uk\u00e1za\u0165 rozsah korel\u00e1cie (ak existuje) medzi hodnotami pozorovan\u00fdch veli\u010d\u00edn alebo javov<\/b> (naz\u00fdvan\u00fdch premenn\u00e9). Ak medzi premenn\u00fdmi neexistuje \u017eiadna korel\u00e1cia, body sa n\u00e1hodne rozpt\u00fdlia v s\u00faradnicovej rovine. Ak existuje ve\u013ek\u00e1 korel\u00e1cia, body sa s\u00fastredia v bl\u00edzkosti priamky. Scatter grafy s\u00fa u\u017eito\u010dn\u00fdmi n\u00e1strojmi na<\/span> vizualiz\u00e1ciu \u00fadajov na ilustr\u00e1ciu trendu<\/b>.<\/span><\/p>\n

    \"\"<\/p>\n

    Ak by sme chceli pou\u017ei\u0165 segment\u00e1ciu v na\u0161om scatterplote, tak pou\u017eijeme n\u00e1sledovn\u00e9:<\/span><\/p>\n

    plotrelseg <- ggplot(gadata, aes(x=sessions, y=transactions, color=deviceCategory)) + geom_point(shape=1)<\/span><\/p>\n

    plot(plotrelseg)<\/span><\/p>\n

    5. Vizualiz\u00e1cia pomocou histogramu<\/h3>\n

    Histogram<\/b> je graf, ktor\u00fd v\u00e1m umo\u017en\u00ed <\/span>objavi\u0165 a zobrazi\u0165 z\u00e1kladn\u00e9 rozdelenie frekvencie<\/b> (tvaru) <\/span>mno\u017einy nepretr\u017eit\u00fdch \u00fadajov<\/b>. To umo\u017e\u0148uje kontrolu \u00fadajov pre jeho z\u00e1kladn\u00fa distrib\u00faciu (napr. norm\u00e1lne rozdelenie), od\u013eahl\u00e9 hodnoty, skewness at\u010f. Pr\u00edklad histogramu a spracovan\u00fdch \u00fadajov, z ktor\u00fdch bol vytvoren\u00fd, je uveden\u00fd ni\u017e\u0161ie:<\/span><\/p>\n

    \"\"<\/p>\n

    ggt <- ggplot(gadata, aes(x=sessions)) + <\/span><\/p>\n

    \u00a0geom_histogram(aes(y=..density..), \u00a0\u00a0\u00a0\u00a0\u00a0# Histogram with density instead of count on y-axis<\/span><\/p>\n

    \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0binwidth=50,<\/span><\/p>\n

    \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0colour=“black“, fill=“white“) +<\/span><\/p>\n

    \u00a0geom_density(alpha=.2, fill=“#FF6666″) \u00a0# Overlay with transparent density plot<\/span><\/p>\n

    plot(ggt)<\/span><\/p>\n

    Histogram je skvel\u00fd graf, kde si <\/span>m\u00f4\u017eete do r\u00f4znych ko\u0161ov rozdeli\u0165 na\u0161e d\u00e1ta<\/b>. A to bu\u010f pod\u013ea va\u0161ich preferenci\u00ed, alebo v\u00fdpo\u010dtu, ko\u013eko ko\u0161ov pou\u017ei\u0165 a v akej ve\u013ekosti. <\/span><\/p>\n

    TIP:<\/b><\/p>\n

    Napr\u00edklad metrika ako Priemern\u00fd \u010das na str\u00e1nke. Je zauj\u00edmav\u00e9 ju analyzova\u0165 (aj ke\u010f m\u00e1 svoje nedostatky), ale v Google Analytics n\u00e1m sama o sebe ve\u013ea nepovie. Ale ak by sme si t\u00fato metriku vizualizovali pomocou histogramu a zatriedili ko\u0161e napr\u00edklad po 2 minutov\u00fdch intervaloch \u00a0tak by sme z\u00edskali omnoho zauj\u00edmavej\u0161ie d\u00e1ta o tom ako dlho u n\u00e1s n\u00e1v\u0161tevn\u00edci str\u00e1via \u010dasu.<\/span><\/p>\n

    Z\u00e1ver<\/span><\/h2>\n

    Predstavili sme si 5 zauj\u00edmav\u00fdch grafov, pomocou ktor\u00fdch viete vizualizova\u0165 va\u0161e d\u00e1ta. Bez oh\u013eadu na to, \u010di s\u00fa z Google Analytics alebo nie, syntax je st\u00e1le rovnak\u00e1. Ak by ste mali nejak\u00e9 zauj\u00edmav\u00e9 n\u00e1pady na vizualiz\u00e1ciu d\u00e1t pomocou R, ur\u010dite nev\u00e1hajte a nap\u00ed\u0161te n\u00e1m koment\u00e1r.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

    V poslednom \u010dl\u00e1nku som p\u00edsal o sp\u00f4sobe, ako z\u00edska\u0165 d\u00e1ta z Google Analytics pomocou R. V dne\u0161nom sa…<\/p>\n","protected":false},"author":66,"featured_media":14193,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[200],"tags":[],"_links":{"self":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/14185"}],"collection":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/users\/66"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/comments?post=14185"}],"version-history":[{"count":6,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/14185\/revisions"}],"predecessor-version":[{"id":14204,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/posts\/14185\/revisions\/14204"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/media\/14193"}],"wp:attachment":[{"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/media?parent=14185"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/categories?post=14185"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.dase-analytics.com\/blog\/sk\/wp-json\/wp\/v2\/tags?post=14185"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}