{"id":53531,"date":"2019-12-04T00:00:00","date_gmt":"2019-12-03T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/ishhem-anomalii-i-predskazyvaem-sboi-s-pomoshhyu-nejrosetej"},"modified":"2020-02-18T14:01:26","modified_gmt":"2020-02-18T11:01:26","slug":"ishhem-anomalii-i-predskazyvaem-sboi-s-pomoshhyu-nejrosetej","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/ishhem-anomalii-i-predskazyvaem-sboi-s-pomoshhyu-nejrosetej","title":{"rendered":"Po k\u00ebrkoni anomali dhe parashikoni defekte me ndihm\u00ebn e rrjeteve nervore","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Po<\/p>\n

<\/p>\n

Zhvillimi i sistemeve softuerike industriale k\u00ebrkon v\u00ebmendje t\u00eb madhe ndaj q\u00ebndrueshm\u00ebris\u00eb s\u00eb produktit p\u00ebrfundimtar, si dhe reagim t\u00eb shpejt\u00eb ndaj d\u00ebshtimeve dhe problemeve, n\u00ebse ato ndodhin. Monitorimi, sigurisht, ndihmon n\u00eb reagimin ndaj d\u00ebshtimeve dhe problemeve n\u00eb m\u00ebnyr\u00eb m\u00eb efikase dhe m\u00eb t\u00eb shpejt\u00eb, por nuk \u00ebsht\u00eb mjaftuesh\u00ebm. S\u00eb pari, \u00ebsht\u00eb e v\u00ebshtir\u00eb t\u00eb mbahen n\u00ebn kontroll nj\u00eb num\u00ebr t\u00eb madh server\u00ebsh \u2013 k\u00ebrkohet nj\u00eb num\u00ebr i madh njer\u00ebzish. S\u00eb dyti, \u00ebsht\u00eb e nevojshme t\u00eb kuptohet mir\u00eb se si funksionon aplikacioni, p\u00ebr t\u00eb parashikuar gjendjen e tij. Prandaj, nevojitet shum\u00eb njer\u00ebz, q\u00eb kuptojn\u00eb mir\u00eb sistemet q\u00eb zhvillojm\u00eb, treguesit e tyre dhe karakteristikat e tyre. T\u00eb supozosh, edhe n\u00ebse gjen nj\u00eb num\u00ebr t\u00eb mjaftuesh\u00ebm njer\u00ebzish t\u00eb gatsh\u00ebm p\u00ebr t\u00eb b\u00ebr\u00eb k\u00ebt\u00eb, k\u00ebrkohet gjithashtu shum\u00eb koh\u00eb p\u00ebr t\u2019i trajnuar ata.<\/p>\n

<\/p>\n

\u00c7far\u00eb duhet b\u00ebr\u00eb? K\u00ebtu na vjen n\u00eb ndihm\u00eb inteligjenca artificiale. Artikulli do t\u00eb flas\u00eb p\u00ebr mir\u00ebmbajtjen parashikuese<\/a><\/noindex> (predictive maintenance). Ky qasje po fiton popullaritet n\u00eb m\u00ebnyr\u00eb aktive. Jan\u00eb shkruar shum\u00eb artikuj, p\u00ebrfshir\u00eb edhe n\u00eb Habr\u00eb. Kompanit\u00eb e m\u00ebdha e shp\u00ebrfaqin n\u00eb p\u00ebrdorim k\u00ebt\u00eb qasje p\u00ebr t\u00eb mbajtur n\u00eb funksion serverat e tyre. Pas studimit t\u00eb nj\u00eb numri t\u00eb madh artikujsh, ne vendos\u00ebm ta provojm\u00eb k\u00ebt\u00eb qasje. \u00c7far\u00eb rezultoi nga kjo? <\/p>\n

<\/a><\/noindex><\/p>\n

Hyrje<\/h2>\n

<\/p>\n

Sistemi i zhvilluar softuerik, her\u00ebt a von\u00eb, del n\u00eb p\u00ebrdorim. \u00cbsht\u00eb e r\u00ebnd\u00ebsishme p\u00ebr p\u00ebrdoruesin q\u00eb sistemi t\u00eb funksionoj\u00eb pa probleme. N\u00ebse ndonj\u00eb situat\u00eb e papritur ndodh, ajo duhet t\u00eb rregullohet me minimumin e vonesave. <\/p>\n

<\/p>\n

P\u00ebr t\u00eb thjeshtuar mb\u00ebshtetje teknike t\u00eb sistemit softuerik, ve\u00e7an\u00ebrisht n\u00ebse ka shum\u00eb server\u00eb, zakonisht p\u00ebrdoren programe monitorimi, t\u00eb cilat marrin metrika nga sistemi softuerik n\u00eb funksionim, ofrojn\u00eb mund\u00ebsin\u00eb p\u00ebr t\u00eb diagnostikuar gjendjen e tij dhe ndihmojn\u00eb n\u00eb p\u00ebrcaktimin e asaj q\u00eb shkaktoi d\u00ebshtimin. Ky proces quhet monitorimi i sistemit softuerik.<\/p>\n

\n\"Po<\/p>\n

Figura 1. Nd\u00ebrfaqja p\u00ebr monitorimin grafana<\/em><\/p>\n

Metrikat jan\u00eb tregues t\u00eb ndrysh\u00ebm t\u00eb sistemit softuerik, mjedisit t\u00eb tij t\u00eb ekzekutimit ose kompjuterit fizik n\u00eb t\u00eb cilin \u00ebsht\u00eb instaluar sistemi me etiket\u00ebn e koh\u00ebs, momentin kur jan\u00eb marr\u00eb metrikat. N\u00eb analiz\u00ebn statike, t\u00eb dh\u00ebnat e metrikave quhen serit\u00eb kohore. P\u00ebr t\u00eb v\u00ebzhguar gjendjen e sistemit softuerik, metrikat shfaqen n\u00eb form\u00ebn e grafik\u00ebve: n\u00eb boshtin X \u2013 koha, dhe n\u00eb boshtin Y \u2013 vlerat (figura 1). Nga nj\u00eb sistem softuerik n\u00eb funksionim mund t\u00eb merren disa mij\u00ebra metrika (nga \u00e7do nod). Ato formojn\u00eb hap\u00ebsir\u00ebn e metrikave (serit\u00eb e kohshme shum\u00ebdimensionale). <\/p>\n

<\/p>\n

Duke qen\u00eb se n\u00eb sistemet softuerike komplekse merren nj\u00eb num\u00ebr i madh metrikash, monitorimi manual b\u00ebhet nj\u00eb detyr\u00eb e v\u00ebshtir\u00eb. P\u00ebr t\u00eb reduktuar v\u00ebllimin e t\u00eb dh\u00ebnave q\u00eb analizohen nga administrator\u00ebt, mjetet e monitorimit p\u00ebrmbajn\u00eb mjete p\u00ebr zbulimin e automatik t\u00eb problemeve t\u00eb mundshme. P\u00ebr shembull, mund t\u00eb konfigurohet nj\u00eb trigger q\u00eb aktivizohet n\u00eb rast se hap\u00ebsira e lir\u00eb diskore zvog\u00eblohet deri n\u00eb nj\u00eb prag t\u00eb caktuar. Gjithashtu mund t\u00eb diagnostikohet automatikisht ndalimi i serverit ose ngadal\u00ebsimi kritik i shpejt\u00ebsis\u00eb s\u00eb sh\u00ebrbimit. N\u00eb praktik\u00eb, mjetet e monitorimit p\u00ebrballen mir\u00eb me zbulimin e d\u00ebshtimeve q\u00eb kan\u00eb ndodhur ose identifikimin e simboleve t\u00eb thjeshta t\u00eb d\u00ebshtimeve t\u00eb ardhshme, por p\u00ebrgjith\u00ebsisht parashikimi i d\u00ebshtimeve t\u00eb mundshme mbetet nj\u00eb gozhd\u00eb e v\u00ebshtir\u00eb p\u00ebr ta. Parashikimi i d\u00ebshtimeve p\u00ebrmes analiz\u00ebs manuale t\u00eb metrikave k\u00ebrkon angazhimin e specialist\u00ebve t\u00eb kualifikuar. Ai \u00ebsht\u00eb me produktivitet t\u00eb ul\u00ebt. shumica e d\u00ebshtimeve t\u00eb mundshme mund t\u00eb kalojn\u00eb pa u v\u00ebn\u00eb re.<\/p>\n

<\/p>\n

N\u00eb vitet e fundit, midis kompanive t\u00eb m\u00ebdha IT q\u00eb zhvillojn\u00eb software, po fiton popullaritet pik\u00ebrisht mir\u00ebmbajtja parashikuese e sistemeve softuerike. Thelbi i k\u00ebtij qasje \u00ebsht\u00eb identifikimi i problemeve q\u00eb \u00e7ojn\u00eb n\u00eb degradimin e sistemit n\u00eb fazat e hershme, para d\u00ebshtimit duke p\u00ebrdorur inteligjenc\u00ebn artificiale. Kjo qasje nuk e p\u00ebrjashtan plot\u00ebsisht monitorimin manual t\u00eb sistemit. Ai \u00ebsht\u00eb ndihm\u00ebs p\u00ebr procesin e monitorimit n\u00eb t\u00ebr\u00ebsi. <\/p>\n

<\/p>\n

Mjeti kryesor p\u00ebr zbatimin e mir\u00ebmbajtjes parashikuese \u00ebsht\u00eb detyra e gjetjes s\u00eb anomalis\u00eb n\u00eb serit\u00eb kohore, pasi n\u00eb rast se ndodh nj\u00eb anomali<\/strong> n\u00eb t\u00eb dh\u00ebna, \u00ebsht\u00eb e mundur q\u00eb pas nj\u00eb kohe t\u00eb ndodh\u00eb nj\u00eb d\u00ebshtim ose keqfunksionim<\/strong>. Anomalia \u00ebsht\u00eb nj\u00eb shk\u00ebputje e caktuar e treguesve t\u00eb sistemit softuerik, si\u00e7 \u00ebsht\u00eb identifikimi i degradimit t\u00eb shpejt\u00ebsis\u00eb s\u00eb ekzekutimit t\u00eb nj\u00eb lloji k\u00ebrkese ose ulja e numrit mesatar t\u00eb k\u00ebrkesave t\u00eb sh\u00ebrbimit me nj\u00eb nivel t\u00eb q\u00ebndruesh\u00ebm t\u00eb sesioneve t\u00eb klient\u00ebve.<\/p>\n

<\/p>\n

Detyra e k\u00ebrkimit t\u00eb anomalive p\u00ebr sistemet software ka specifikat e saj. Idealisht, p\u00ebr \u00e7do sistem software \u00ebsht\u00eb e nevojshme zhvillimi ose p\u00ebrmir\u00ebsimi i metodave ekzistuese, pasi k\u00ebrkimi i anomalive varet shum\u00eb nga t\u00eb dh\u00ebnat n\u00eb t\u00eb cilat kryhet, nd\u00ebrsa t\u00eb dh\u00ebnat e sistemeve software ndryshojn\u00eb ndjesh\u00ebm n\u00eb var\u00ebsi t\u00eb mjeteve t\u00eb zbatimit, deri n\u00eb at\u00eb pik\u00eb sa n\u00ebn sistemin n\u00ebn t\u00eb cilin \u00ebsht\u00eb nisur.<\/p>\n

<\/p>\n

Metodat e k\u00ebrkimit t\u00eb anomalive n\u00eb parashikimin e d\u00ebshtimeve t\u00eb sistemeve software<\/h2>\n

<\/p>\n

Para s\u00eb gjithash, duhet th\u00ebn\u00eb se ideja e parashikimit t\u00eb d\u00ebshtimeve u frym\u00ebzua nga artikulli \u00abM\u00ebsimi i makinerive n\u00eb monitorimin IT\u00bb<\/a><\/noindex>. P\u00ebr t\u00eb verifikuar efektivitetin e qasjes me k\u00ebrkimin automatik t\u00eb anomalive, u zgjodh sistemi software \u00abWeb-Konsolidimi\u00bb, i cili \u00ebsht\u00eb nj\u00eb prej projekteve t\u00eb kompanis\u00eb NPO \u00abKrista\u00bb. P\u00ebr t\u00eb, m\u00eb par\u00eb ishin kryer monitorime manuale sipas metrikeve t\u00eb marra. Duke qen\u00eb se sistemi \u00ebsht\u00eb mjaft i komplikuar, p\u00ebr t\u00eb merren nj\u00eb num\u00ebr t\u00eb madh metrikeve: tregues t\u00eb JVM (ngarkesa e mbledh\u00ebsit t\u00eb plehrave), tregues t\u00eb OS-s\u00eb n\u00ebn t\u00eb cil\u00ebn ekzekutohet kodi (memoria virtuale, % ngarkesa e CPU-s\u00eb), tregues t\u00eb rrjetit (ngarkesa e rrjetit), t\u00eb serverit vet\u00eb (ngarkesa e CPU-s\u00eb, memorjes), metrike t\u00eb wildfly dhe metrike t\u00eb veta t\u00eb aplikacionit p\u00ebr t\u00eb gjitha pjes\u00ebt kritike t\u00eb sistemit. <\/p>\n

<\/p>\n

T\u00eb gjitha metrike merren nga sistemi p\u00ebrmes graphite. Fillimisht, u p\u00ebrdor baza whisper si zgjidhje standarde p\u00ebr grafan\u00ebn, por me rritjen e baz\u00ebs s\u00eb klient\u00ebve, graphite nuk arriti m\u00eb t\u00eb p\u00ebrballoj\u00eb ngarkes\u00ebn e kapacitetit t\u00eb n\u00ebn-sistemit t\u00eb diskut t\u00eb DC-s\u00eb. Pas k\u00ebsaj, u mor vendimi p\u00ebr t\u00eb k\u00ebrkuar nj\u00eb zgjidhje m\u00eb efikase. Zgjedhja ra n\u00eb favor t\u00eb graphite+clickhouse<\/a><\/noindex>, e cila lejoi uljen e ngarkes\u00ebs n\u00eb n\u00ebn-sistemin e diskut me nj\u00eb rend dhe pes\u00eb deri n\u00eb gjasht\u00eb her\u00eb t\u00eb zvog\u00ebloj\u00eb hap\u00ebsir\u00ebn e z\u00ebn\u00eb n\u00eb disk. M\u00eb posht\u00eb \u00ebsht\u00eb paraqitur diagrami i mekanizmit t\u00eb mbledhjes s\u00eb metrikeve duke p\u00ebrdorur graphite+clickhouse (figura 2).<\/p>\n

\n\"Po
\n<\/p>\n

Figura 2. Diagrami i marrjes s\u00eb metrikeve<\/em><\/p>\n

Diagrama \u00ebsht\u00eb marr\u00eb nga dokumentacioni i brendsh\u00ebm. Ajo tregon shk\u00ebmbimin e t\u00eb dh\u00ebnave midis grafan\u00ebs (nd\u00ebrfaqja p\u00ebrdoruese p\u00ebr monitorimin q\u00eb p\u00ebrdorim) dhe graphite. Marrja e metrikeve nga aplikacioni realizohet nga nj\u00eb program i ve\u00e7ant\u00eb \u2013 jmxtrans<\/a><\/noindex>. Ai gjithashtu i grumbullon ato n\u00eb graphite.
\nSistemi \u00abWeb-Konsolidimi\u00bb ka disa karakteristika q\u00eb krijojn\u00eb probleme p\u00ebr parashikimin e d\u00ebshtimeve:<\/p>\n

<\/p>\n

    \n
  1. trendet ndodhin shpesh t\u00eb ndryshojn\u00eb. P\u00ebr k\u00ebt\u00eb sistem software, l\u00ebshohen versione t\u00eb ndryshme. \u00c7do nj\u00ebra nga ato sjell ndryshime n\u00eb pjes\u00ebn software t\u00eb sistemit. Sipas k\u00ebsaj, zhvilluesit ndikojn\u00eb drejtp\u00ebrdrejt n\u00eb metrike t\u00eb k\u00ebtij sistemi dhe mund t\u00eb shkaktojn\u00eb ndryshimin e trendit; <\/li>\n
  2. karakteristika e zbatimit dhe gjithashtu q\u00ebllimet e p\u00ebrdorimit nga klient\u00ebt e k\u00ebtij sistemi shpesh shkaktojn\u00eb anomalit\u00eb pa degradim t\u00eb m\u00ebparsh\u00ebm; <\/li>\n
  3. pjesa e anomalive n\u00eb raport me t\u00eb gjith\u00eb grupin e t\u00eb dh\u00ebnave \u00ebsht\u00eb e vog\u00ebl (< 5%); <\/li>\n
  4. mund t\u00eb ndodhin nd\u00ebrprerje n\u00eb marrjen e treguesve nga sistemi. N\u00eb disa periudha t\u00eb shkurtra, sistemi i monitorimit nuk arrin t\u00eb marr\u00eb metrike. P\u00ebr shembull, n\u00ebse serveri \u00ebsht\u00eb i ngarkuar. P\u00ebr trajnimin e rrjeteve nervore, kjo \u00ebsht\u00eb kritike. Ndodhet nevoja p\u00ebr t\u00eb plot\u00ebsuar boshll\u00ebqet n\u00eb m\u00ebnyr\u00eb sintetike;<\/li>\n
  5. Rastet me anomalit\u00eb shpesh jan\u00eb relevante vet\u00ebm p\u00ebr nj\u00eb num\u00ebr \/ muaji \/ koh\u00eb t\u00eb caktuar (sezonalitet). Ky sistem ka nj\u00eb rregull t\u00eb qart\u00eb t\u00eb p\u00ebrdorimit nga p\u00ebrdoruesit e tij. Si rezultat, metrike jan\u00eb relevante vet\u00ebm p\u00ebr nj\u00eb periudh\u00eb t\u00eb caktuar. Sistemi mund t\u00eb p\u00ebrdoret jo vazhdimisht, por vet\u00ebm n\u00eb disa muaj: p\u00ebrzgjedhur n\u00eb var\u00ebsi t\u00eb vitit. Ndodhin situata kur e nj\u00ebjta sjellje e metrikeve n\u00eb nj\u00eb rast mund t\u00eb \u00e7oj\u00eb n\u00eb d\u00ebshtimin e sistemit software, nd\u00ebrsa n\u00eb nj\u00eb rast tjet\u00ebr jo.
    \nFillimisht u analizuan metodat e zbulimit t\u00eb anomalive n\u00eb t\u00eb dh\u00ebnat e monitorimit t\u00eb sistemeve software. N\u00eb artikujt mbi k\u00ebt\u00eb tem\u00eb, n\u00eb p\u00ebrqindje t\u00eb vogla anomalish n\u00eb raport me grupin tjet\u00ebr t\u00eb dh\u00ebnash, shpesh sugjerohet p\u00ebrdorimi i rrjeteve nervore. <\/li>\n<\/ol>\n

    <\/p>\n

    Logjika kryesore p\u00ebr k\u00ebrkimin e anomalive me ndihm\u00ebn e t\u00eb dh\u00ebnave t\u00eb rrjeteve nervore \u00ebsht\u00eb paraqitur n\u00eb figur\u00ebn 3:<\/p>\n

    \n\"Po
    \n<\/p>\n

    Figura 3. K\u00ebrkimi i anomalive me ndihm\u00ebn e rrjetit nervor<\/em><\/p>\n

    N\u00eb rezultatin e parashikimit ose rikthimin e dritares aktuale t\u00eb rrjedh\u00ebs s\u00eb metrikeve, llogaritet devijimi nga t\u00eb dh\u00ebnat e marra nga sistemi software n\u00eb pun\u00eb. N\u00eb rast se ka nj\u00eb diferenc\u00eb t\u00eb madhe midis trek\u00ebnd\u00ebshave t\u00eb marra nga sistemi software dhe rrjetit nervor, mund t\u00eb konkludohet se segmenti aktual i t\u00eb dh\u00ebnave \u00ebsht\u00eb anomali. Ndodhin disa probleme p\u00ebr p\u00ebrdorimin e rrjeteve nervore:<\/p>\n

    <\/p>\n

      \n
    1. p\u00ebr funksionimin e sakt\u00eb n\u00eb m\u00ebnyr\u00eb t\u00eb rrjedhshme, t\u00eb dh\u00ebnat p\u00ebr trajnimin e modeleve t\u00eb rrjeteve nervore duhet t\u00eb p\u00ebrfshijn\u00eb vet\u00ebm t\u00eb dh\u00ebna \u00abnormale\u00bb; <\/li>\n
    2. \u00ebsht\u00eb e nevojshme t\u00eb kesh nj\u00eb model aktual p\u00ebr t\u00eb b\u00ebr\u00eb zbulime t\u00eb sakta. Ndryshimi i trendit dhe sezonalitetit n\u00eb metrika mund t\u00eb shkaktoj\u00eb nj\u00eb num\u00ebr t\u00eb madh t\u00eb sinjaleve false t\u00eb modelit. P\u00ebr ta p\u00ebrdit\u00ebsuar at\u00eb, \u00ebsht\u00eb e domosdoshme t\u00eb p\u00ebrcaktohet qart\u00eb koha kur modeli \u00ebsht\u00eb b\u00ebr\u00eb i vjet\u00ebruar. N\u00ebse modeli p\u00ebrdit\u00ebsohet shum\u00eb her\u00ebt ose shum\u00eb von\u00eb, probabiliteti i sinjaleve false do t\u00eb jet\u00eb i lart\u00eb.
      \nGjithashtu, nuk duhet harruar k\u00ebrkimin dhe parandalimin e shfaqjes s\u00eb shpesht\u00eb t\u00eb sinjaleve false. Ato pritet t\u00eb ndodhin m\u00eb s\u00eb shpeshti n\u00eb situata t\u00eb jashtzakonshme. Megjithat\u00eb, ato mund t\u00eb jen\u00eb gjithashtu pasoj\u00eb e nj\u00eb gabimi n\u00eb rrjetin nervor p\u00ebr shkak t\u00eb m\u00ebsimit t\u00eb pamjaftuesh\u00ebm. Duhet t\u00eb minimizohet numri i sinjaleve false t\u00eb modelit. N\u00eb t\u00eb kund\u00ebrt, parashikimet e gabuara do t\u00eb shpenzojn\u00eb shum\u00eb koh\u00eb t\u00eb administratorit, e destinuar p\u00ebr t\u00eb verifikuar sistemin. M\u00eb von\u00eb ose m\u00eb von\u00eb, do t\u00eb ndodh\u00eb q\u00eb administratorit t\u00eb ndaloj\u00eb s\u00eb reaguar ndaj nj\u00eb sistemi monitorimi 'paranojak'.<\/li>\n<\/ol>\n

      <\/p>\n

      Rrjeti nervor rekurent<\/h2>\n

      <\/p>\n

      P\u00ebr zbulimin e anomalive n\u00eb seri kohore, mund t\u00eb p\u00ebrdoret rrjeti nervor rekurent <\/a><\/noindex>me memorie LSTM. Problemi q\u00ebndron vet\u00ebm n\u00eb faktin se ai mund t\u00eb p\u00ebrdoret vet\u00ebm p\u00ebr seri kohore q\u00eb parashikohen. N\u00eb rastin ton\u00eb, jo t\u00eb gjitha metrikat jan\u00eb t\u00eb parashikueshme. P\u00ebrpjekja p\u00ebr t\u00eb aplikuar RNN LSTM p\u00ebr nj\u00eb seri kohore \u00ebsht\u00eb paraqitur n\u00eb figur\u00ebn 4.<\/p>\n

      \n\"Po
      \n<\/p>\n

      Figura 4. Shembulli i funksionimit t\u00eb rrjetit nervor rekurent me qeliza memorie LSTM.<\/em><\/p>\n

      Si\u00e7 duket nga figura 4, RNN LSTM arriti t\u00eb diagnozoj\u00eb anomalin\u00eb n\u00eb k\u00ebt\u00eb segment t\u00eb koh\u00ebs. Aty ku rezultati ka nj\u00eb gabim t\u00eb lart\u00eb parashikimi (gabimi mesatar), me t\u00eb v\u00ebrtet\u00eb ndodhi nj\u00eb anomalie n\u00eb tregues. P\u00ebrdorimi i vet\u00ebm t\u00eb RNN LSTM do t\u00eb jet\u00eb padyshim i pamjaftuesh\u00ebm, pasi \u00ebsht\u00eb i aplikuesh\u00ebm p\u00ebr nj\u00eb num\u00ebr t\u00eb vog\u00ebl metrikash. Mund t\u00eb p\u00ebrdoret si nj\u00eb metod\u00eb ndihmuese p\u00ebr zbulimin e anomalive. <\/p>\n

      <\/p>\n

      Auto-koduesi p\u00ebr parashikimin e d\u00ebshtimeve<\/h2>\n

      <\/p>\n

      Auto-koduesi<\/a><\/noindex> \u2013 n\u00eb thelb \u00ebsht\u00eb nj\u00eb rrjet nervor artificial. Shtresa hyr\u00ebse \u00ebsht\u00eb encoder, nd\u00ebrsa shtresa dal\u00ebse \u00ebsht\u00eb decoder. Disavantazhi i t\u00eb gjitha rrjeteve nervore t\u00eb k\u00ebtij lloji \u00ebsht\u00eb se ato lokalizojn\u00eb ndryshe anomalit\u00eb. U zgjodh arkitektura e auto-koduesit t\u00eb sinkronizuar.<\/p>\n

      \n\"Po
      \n<\/p>\n

      Figura 5. Shembulli i funksionimit t\u00eb auto-koduesit.<\/em><\/p>\n

      Auto-koduesit trajnohen n\u00eb t\u00eb dh\u00ebna normale dhe m\u00eb pas gjejn\u00eb di\u00e7ka anomale n\u00eb t\u00eb dh\u00ebnat e futur n\u00eb model. Kjo \u00ebsht\u00eb pik\u00ebrisht ajo q\u00eb na nevojitet p\u00ebr k\u00ebt\u00eb detyr\u00eb. Tani mbetet vet\u00ebm t\u00eb zgjidhni se cili nga auto-koduesit do t\u00eb p\u00ebrshtatet m\u00eb mir\u00eb p\u00ebr k\u00ebt\u00eb detyr\u00eb. Forma m\u00eb e thjesht\u00eb arkitektonike e auto-koduesit \u00ebsht\u00eb nj\u00eb rrjet nervor i drejtp\u00ebrdrejt\u00eb, jo t\u00eb rikthyesh\u00ebm, q\u00eb \u00ebsht\u00eb shum\u00eb e ngjashme me perceptronin shum\u00eb t\u00eb shtresave<\/a><\/noindex> (multilayer perceptron, MLP), me nivelin hyr\u00ebs, nivelin dal\u00ebs dhe nj\u00eb ose disa shtresa t\u00eb fshehura q\u00eb i lidhin ato.
      \nMegjithat\u00eb, dallimi midis auto-koduesve dhe MLP q\u00ebndron n\u00eb faktin se n\u00eb auto-koduesin niveli i daljes ka t\u00eb nj\u00ebjtin num\u00ebr nyjesh q\u00eb ka niveli hyr\u00ebs, dhe q\u00eb n\u00eb vend t\u00eb m\u00ebsimit p\u00ebr t\u00eb parashikuar vler\u00ebn e synuar Y, e cila i atribuohet hyrjes X, auto-koduesi m\u00ebson t\u00eb rikonstruktoj\u00eb vet\u00eb X. Prandaj, auto-koduesit jan\u00eb modele t\u00eb t\u00eb m\u00ebsuarit t\u00eb pakontrolluar. <\/p>\n

      <\/p>\n

      Detyra e auto-koduesit \u00ebsht\u00eb t\u00eb gjej\u00eb indekset kohore r0\u2026rn, t\u00eb cilat p\u00ebrkojn\u00eb me elementet anomale n\u00eb vektorin hyr\u00ebs X. Ky efekt arrihet p\u00ebrmes k\u00ebrkimit t\u00eb gabimit katror.<\/p>\n

      \n\"Po <\/p>\n

      Figura 6. Auto-koduesi sinkron.<\/em><\/p>\n

      P\u00ebr auto-koduesin, u zgjodh arkitektura sinkronike<\/a><\/noindex>. Avantazhet e saj: mund\u00ebsia e p\u00ebrdorimit t\u00eb nj\u00eb mode t\u00eb p\u00ebrpunimit n\u00eb fluks dhe nj\u00eb num\u00ebr relativisht m\u00eb t\u00eb vog\u00ebl parametrash t\u00eb rrjetit nervor n\u00eb krahasim me arkitektura t\u00eb tjera.<\/p>\n

      <\/p>\n

      Mekanizmi i minimizimit t\u00eb sinjaleve false<\/h2>\n

      <\/p>\n

      Duke pasur parasysh se ndodhin situata t\u00eb ndryshme jasht\u00ebzakonisht, si dhe mund t\u00eb ndodh\u00eb situata e m\u00ebsimit t\u00eb pamjaftuesh\u00ebm t\u00eb rrjetit nervor, p\u00ebr modelin e zhvilluar t\u00eb zbulimit t\u00eb anomalive, u mor nj\u00eb vendim p\u00ebr nevoj\u00ebn e zhvillimit t\u00eb nj\u00eb mekanizmi p\u00ebr minimizimin e sinjaleve false. Ky mekaniz\u00ebm bazohet n\u00eb nj\u00eb baz\u00eb modelesh, q\u00eb e klasifikon administratori. <\/p>\n

      <\/p>\n

      Algoritmi i transformimit dinamik t\u00eb vij\u00ebs kohore<\/a><\/noindex> (algoritmi DTW, nga anglishtja dynamic time warping) lejon t\u00eb gjej\u00eb p\u00ebrputhjen m\u00eb t\u00eb optimizuar midis sekuencave t\u00eb koh\u00ebs. I pari u aplikua n\u00eb njohjen e t\u00eb folurit: u p\u00ebrdor p\u00ebr t\u00eb p\u00ebrcaktuar se si dy sinjale t\u00eb t\u00eb folurit p\u00ebrfaq\u00ebsojn\u00eb t\u00eb nj\u00ebjtin fraz\u00eb t\u00eb shqiptuar.<\/p>\n

      <\/p>\n

      Parimi kryesor i minimizimit t\u00eb alarmeve t\u00eb rreme \u00ebsht\u00eb grumbullimi i nj\u00eb baze referencash me ndihm\u00ebn e nj\u00eb operatori, i cili klasifikon rastet e dyshimta, t\u00eb zbuluara nga rrjetet nervore. M\u00eb pas, b\u00ebhet nj\u00eb krahasim i standardit t\u00eb klasifikuar me rastin q\u00eb zbuloi sistemi, dhe merret nj\u00eb p\u00ebrfundim n\u00ebse rasti \u00ebsht\u00eb nj\u00eb alarm i rrem\u00eb apo nj\u00eb q\u00eb \u00e7on n\u00eb d\u00ebshtim. Pikerisht p\u00ebr krahasimin e dy serive t\u00eb koh\u00ebs p\u00ebrdoret algoritmi DTW. Instrumenti kryesor p\u00ebr minimizimin, megjithat\u00eb, \u00ebsht\u00eb klasifikimi. Supozohet se pas grumbullimit t\u00eb nj\u00eb numri t\u00eb madh t\u00eb rast\u00ebve t\u00eb standardeve, sistemi do t\u00eb filloj\u00eb t\u00eb pyes\u00eb m\u00eb pak operatorin p\u00ebr shkak t\u00eb ngjashm\u00ebris\u00eb s\u00eb shumic\u00ebs s\u00eb rasteve dhe shfaqjes s\u00eb situatave t\u00eb ngjashme.<\/p>\n

      <\/p>\n

      Si rezultat i metodave t\u00eb m\u00ebsip\u00ebrme, \u00ebsht\u00eb nd\u00ebrtuar nj\u00eb program eksperimental p\u00ebr parashikimin e d\u00ebshtimeve t\u00eb sistemit \"Web-Konsolidimi\". Q\u00ebllimi i k\u00ebtij programi ishte, duke p\u00ebrdorur arkivin ekzistues t\u00eb t\u00eb dh\u00ebnave t\u00eb monitorimit dhe informacionin mbi d\u00ebshtimet e ndodhura, t\u00eb vler\u00ebsoj\u00eb aft\u00ebsin\u00eb e k\u00ebtij qasje p\u00ebr sistemet tona softuerike. Skema e funksionimit t\u00eb programit \u00ebsht\u00eb paraqitur m\u00eb posht\u00eb, n\u00eb figur\u00ebn 7.<\/p>\n

      \n\"Po <\/p>\n

      Figura 7. Skema e parashikimit t\u00eb d\u00ebshtimeve mbi baz\u00ebn e analiz\u00ebs s\u00eb hap\u00ebsir\u00ebs s\u00eb metrikave.<\/em><\/p>\n

      N\u00eb skem\u00eb mund t\u00eb dallohet dy blloqe kryesore: k\u00ebrkimi i segmenteve anomale t\u00eb koh\u00ebs n\u00eb fluxin e t\u00eb dh\u00ebnave t\u00eb monitorimit (metrikave) dhe mekanizmi p\u00ebr minimizimin e alarmeve t\u00eb rreme. Sh\u00ebnim: p\u00ebr q\u00ebllime eksperimentale, t\u00eb dh\u00ebnat merren p\u00ebrmes lidhjes JDBC nga baza e t\u00eb dh\u00ebnave, n\u00eb t\u00eb cil\u00ebn ato ruhen nga graphite.
      \nM\u00eb posht\u00eb \u00ebsht\u00eb paraqitur nd\u00ebrfaqja rezultante t\u00eb sistemit t\u00eb monitorimit t\u00eb zhvilluar (figura 8).<\/p>\n

      \n\"Po<\/p>\n

      Figura 8. Nd\u00ebrfaqja e sistemit eksperimental t\u00eb monitorimit.<\/em><\/p>\n

      N\u00eb nd\u00ebrfaqe shfaqet p\u00ebrqindja e anomalis\u00eb sipas metrikave t\u00eb marra. N\u00eb rastin ton\u00eb, marrja simulohet. Ne tashm\u00eb kemi t\u00eb dh\u00ebna p\u00ebr disa jav\u00eb dhe i ngarkojm\u00eb ato gradualisht p\u00ebr t\u00eb kontrolluar rastin me anomali q\u00eb \u00e7on n\u00eb d\u00ebshtim. N\u00eb status barin n\u00eb fund shfaqet p\u00ebrcenti total i anomalis\u00eb s\u00eb t\u00eb dh\u00ebnave n\u00eb momentin e tanish\u00ebm, q\u00eb p\u00ebrcaktohet me ndihm\u00ebn e auto-shkoduesit. Gjithashtu p\u00ebr metrikat e parashikuara shfaqet nj\u00eb p\u00ebrqindje e ve\u00e7ant\u00eb, e cila llogaritet nga RNN LSTM.<\/p>\n

      <\/p>\n

      Shembull i zbulimit t\u00eb anomalis\u00eb sipas treguesve CPU me ndihm\u00ebn e rrjetit nervor RNN LSTM (figura 9).<\/p>\n

      \n\"Po<\/p>\n

      Figura 9. Zbulimi RNN LSTM.<\/em><\/p>\n

      Nj\u00eb rast mjaft i thjesht\u00eb, n\u00eb thelb nj\u00eb shp\u00ebrthim i zakonsh\u00ebm, megjithat\u00eb q\u00eb \u00e7on n\u00eb d\u00ebshtimin e sistemit, u p\u00ebrcaktua me sukses me ndihm\u00ebn e RNN LSTM. Treguesi i anomalis\u00eb n\u00eb k\u00ebt\u00eb segment kohe \u00ebsht\u00eb 85 \u2013 95%, gjith\u00e7ka mbi 80% (kambani p\u00ebrcaktohet eksperimentalisht) konsiderohet si anomali.
      \nShembull i zbulimit t\u00eb anomalis\u00eb kur sistemi nuk arriti t\u00eb ngarkohet pas nj\u00eb p\u00ebrdit\u00ebsimi. Kjo situat\u00eb detektohet nga auto-shkoduesi (figura 10).<\/p>\n

      \n\"Po<\/p>\n

      Figura 10. Shembuj i zbulimit nga auto-shkoduesi.<\/em><\/p>\n

      Si\u00e7 shihet nga figura, PermGen u ngat\u00ebrrua n\u00eb nj\u00eb nivel. Auto-shkoduesi e konsideroi k\u00ebt\u00eb si t\u00eb \u00e7uditsh\u00ebm, sepse m\u00eb par\u00eb nuk kishte par\u00eb asgj\u00eb t\u00eb till\u00eb. K\u00ebtu anomalia mbetet e 100% deri sa sistemi t\u00eb kthehet n\u00eb gjendje pune. Anomalia shfaqet p\u00ebr t\u00eb gjitha metrikat. Si\u00e7 \u00ebsht\u00eb th\u00ebn\u00eb m\u00eb par\u00eb, auto-shkoduesi nuk di t\u00eb lokalizoj\u00eb anomali. Operator\u00ebt jan\u00eb t\u00eb thirrur p\u00ebr t\u00eb kryer k\u00ebt\u00eb funksion n\u00eb k\u00ebto situata.<\/p>\n

      <\/p>\n

      P\u00ebrfundimi<\/h2>\n

      <\/p>\n

      PC \"Web-Konsolidimi\" \u00ebsht\u00eb n\u00eb zhvillim p\u00ebr disa vite. Sistemi \u00ebsht\u00eb n\u00eb nj\u00eb gjendje mjaft t\u00eb q\u00ebndrueshme dhe numri i incidenteve t\u00eb regjistruara \u00ebsht\u00eb i vog\u00ebl. Megjithat\u00eb, \u00ebsht\u00eb arritur t\u00eb identifikohen anomali q\u00eb \u00e7ojn\u00eb n\u00eb d\u00ebshtim nga 5 \u2013 10 minuta para ndodhjes s\u00eb d\u00ebshtimit. N\u00eb disa raste, njoftimi p\u00ebr d\u00ebshtimin paraprakisht do kishte ndihmuar n\u00eb ruajtjen e koh\u00ebs s\u00eb rregullt, e cila caktohet p\u00ebr kryerjen e pun\u00ebve \"t\u00eb riparimit\".<\/p>\n

      <\/p>\n

      Bazuar n\u00eb eksperimentet q\u00eb jan\u00eb kryer, \u00ebsht\u00eb ende her\u00ebt p\u00ebr t\u00eb b\u00ebr\u00eb p\u00ebrfundime p\u00ebrfundimtare. Aktualisht rezultatet jan\u00eb t\u00eb kund\u00ebrta. N\u00eb nj\u00ebr\u00ebn an\u00eb, duket se algoritmet e bazuara n\u00eb rrjetet nervore jan\u00eb n\u00eb gjendje t\u00eb gjejn\u00eb anomali \"t\u00eb dobishme\". N\u00eb an\u00ebn tjet\u00ebr, mbetet nj\u00eb p\u00ebrqindje e madhe e alarmeve t\u00eb rreme, dhe jo t\u00eb gjitha anomali t\u00eb zbuluara nga specialist\u00eb t\u00eb kualifikuar arrijn\u00eb t\u00eb identifikohen nga rrjeti nervor. Nj\u00eb nga disavantazhet \u00ebsht\u00eb se tani rrjeti nervor k\u00ebrkon m\u00ebsim me m\u00ebsues p\u00ebr t\u00eb punuar normalisht.<\/p>\n

      <\/p>\n

      P\u00ebr t\u00eb avancuar sistemin e parashikimit t\u00eb d\u00ebshtimeve dhe p\u00ebr ta sjell\u00eb n\u00eb nj\u00eb gjendje t\u00eb k\u00ebnaqshme, mund t\u00eb parashikohen disa rrug\u00eb. Kjo \u00ebsht\u00eb nj\u00eb analiz\u00eb m\u00eb e detajuar e rasteve me anomali q\u00eb \u00e7ojn\u00eb n\u00eb d\u00ebshtim, p\u00ebrmes k\u00ebsaj shtese t\u00eb list\u00ebs s\u00eb metrikave t\u00eb r\u00ebnd\u00ebsishme q\u00eb ndikojn\u00eb shum\u00eb n\u00eb gjendjen e sistemit, si dhe heqja e atyre t\u00eb panevojshme q\u00eb nuk e ndikojn\u00eb at\u00eb. Gjithashtu, n\u00ebse e ndjekim k\u00ebt\u00eb drejtim, mund t\u00eb b\u00ebjm\u00eb p\u00ebrpjekje p\u00ebr specializimin e algoritmeve specifikisht p\u00ebr rastet tona me anomali q\u00eb \u00e7ojn\u00eb n\u00eb d\u00ebshtime. Ka dhe nj\u00eb rrug\u00eb tjet\u00ebr. Kjo \u00ebsht\u00eb p\u00ebrmir\u00ebsimi i arkitekturave t\u00eb rrjeteve neurale dhe rritja e sakt\u00ebsis\u00eb s\u00eb zbulimeve p\u00ebrmes k\u00ebsaj, duke shkurtuar koh\u00ebn e trajnimet.<\/p>\n

      <\/p>\n

      Shpreh fal\u00ebnderimin tim ndaj koleg\u00ebve q\u00eb ndihmuan n\u00eb shkrimin dhe ruajtjen e aktualitetit t\u00eb k\u00ebtij artikulli: Viktor Verbitsky<\/a><\/noindex> dhe Sergey Finogenov.<\/p>\n

      Burimi: habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

      \u041f\u0440\u043e\u043c\u044b\u0448\u043b\u0435\u043d\u043d\u0430\u044f \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0430 \u043f\u0440\u043e\u0433\u0440\u0430\u043c\u043c\u043d\u044b\u0445 \u0441\u0438\u0441\u0442\u0435\u043c \u0442\u0440\u0435\u0431\u0443\u0435\u0442 \u0431\u043e\u043b\u044c\u0448\u043e\u0433\u043e \u0432\u043d\u0438\u043c\u0430\u043d\u0438\u044f \u043a \u043e\u0442\u043a\u0430\u0437\u043e\u0443\u0441\u0442\u043e\u0439\u0447\u0438\u0432\u043e\u0441\u0442\u0438 \u043a\u043e\u043d\u0435\u0447\u043d\u043e\u0433\u043e \u043f\u0440\u043e\u0434\u0443\u043a\u0442\u0430, \u0430 \u0442\u0430\u043a\u0436\u0435 \u0431\u044b\u0441\u0442\u0440\u043e\u0433\u043e \u0440\u0435\u0430\u0433\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u044f \u043d\u0430 \u043e\u0442\u043a\u0430\u0437\u044b \u0438 \u0441\u0431\u043e\u0438, \u0435\u0441\u043b\u0438 \u043e\u043d\u0438 \u0432\u0441\u0435-\u0442\u0430\u043a\u0438 \u0441\u043b\u0443\u0447\u0430\u044e\u0442\u0441\u044f. \u041c\u043e\u043d\u0438\u0442\u043e\u0440\u0438\u043d\u0433, \u043a\u043e\u043d\u0435\u0447\u043d\u043e \u0436\u0435, \u043f\u043e\u043c\u043e\u0433\u0430\u0435\u0442 \u0440\u0435\u0430\u0433\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u043d\u0430 \u043e\u0442\u043a\u0430\u0437\u044b \u0438 \u0441\u0431\u043e\u0438 \u044d\u0444\u0444\u0435\u043a\u0442\u0438\u0432\u043d\u0435\u0435 \u0438 \u0431\u044b\u0441\u0442\u0440\u0435\u0435, \u043d\u043e \u043d\u0435\u0434\u043e\u0441\u0442\u0430\u0442\u043e\u0447\u043d\u043e. \u0412\u043e-\u043f\u0435\u0440\u0432\u044b\u0445, \u043e\u0447\u0435\u043d\u044c \u0441\u043b\u043e\u0436\u043d\u043e \u0443\u0441\u043b\u0435\u0434\u0438\u0442\u044c \u0437\u0430 \u0431\u043e\u043b\u044c\u0448\u0438\u043c \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u043e\u043c \u0441\u0435\u0440\u0432\u0435\u0440\u043e\u0432 \u2013 \u043d\u0435\u043e\u0431\u0445\u043e\u0434\u0438\u043c\u043e \u0431\u043e\u043b\u044c\u0448\u043e\u0435 \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u043e \u043b\u044e\u0434\u0435\u0439. \u0412\u043e-\u0432\u0442\u043e\u0440\u044b\u0445, \u043d\u0443\u0436\u043d\u043e \u0445\u043e\u0440\u043e\u0448\u043e \u043f\u043e\u043d\u0438\u043c\u0430\u0442\u044c, \u043a\u0430\u043a […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-53531","post","type-post","status-publish","format-standard","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47K\u00ebrkojm\u00eb anomali dhe parashikojm\u00eb d\u00ebshtime me ndihm\u00ebn e rrjeteve neurale | ProHoster","description":"Zhvillimi industrial.","canonical_url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/ishhem-anomalii-i-predskazyvaem-sboi-s-pomoshhyu-nejrosetej","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"sq_AL","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u0418\u0449\u0435\u043c \u0430\u043d\u043e\u043c\u0430\u043b\u0438\u0438 \u0438 \u043f\u0440\u0435\u0434\u0441\u043a\u0430\u0437\u044b\u0432\u0430\u0435\u043c \u0441\u0431\u043e\u0438 \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e \u043d\u0435\u0439\u0440\u043e\u0441\u0435\u0442\u0435\u0439 | ProHoster","og:description":"\u041f\u0440\u043e\u043c\u044b\u0448\u043b\u0435\u043d\u043d\u0430\u044f \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0430.","og:url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/ishhem-anomalii-i-predskazyvaem-sboi-s-pomoshhyu-nejrosetej","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-12-03T21:00:00+00:00","article:modified_time":"2020-02-18T11:01:26+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"53531","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-24 07:46:24","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-02-28 20:23:55","updated":"2026-01-24 07:46:24","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/53531","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/comments?post=53531"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/53531\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media?parent=53531"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/categories?post=53531"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/tags?post=53531"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}