{"id":97051,"date":"2020-10-15T14:43:34","date_gmt":"2020-10-15T12:43:34","guid":{"rendered":"https:\/\/prohoster.info\/blog\/administrirovanie\/kak-raspoznat-sharlatana-ot-data-science"},"modified":"2020-10-15T14:43:34","modified_gmt":"2020-10-15T12:43:34","slug":"kak-raspoznat-sharlatana-ot-data-science","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/kak-raspoznat-sharlatana-ot-data-science","title":{"rendered":"Si si p\u00ebr t\u00eb identifikuar nj\u00eb mashtrues n\u00eb Data Science?","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Si<\/a><\/noindex>
\nMund t\u00eb keni d\u00ebgjuar p\u00ebr analist\u00eb, specialist\u00eb n\u00eb t\u00eb m\u00ebsuarit e makineris\u00eb dhe inteligjenc\u00ebn artificiale, por a keni d\u00ebgjuar ndonj\u00ebher\u00eb p\u00ebr ata q\u00eb paguhen m\u00eb shum\u00eb se meritojn\u00eb? Takoni mashtruesin e t\u00eb dh\u00ebnave<\/strong>! K\u00ebta mjeshtra, t\u00eb joshur nga puna p\u00ebrfitimprur\u00ebse, krijojn\u00eb nj\u00eb reputacion t\u00eb keq p\u00ebr profesionist\u00ebt e v\u00ebrtet\u00eb t\u00eb p\u00ebrpunimit t\u00eb t\u00eb dh\u00ebnave. N\u00eb k\u00ebt\u00eb material, shqyrtojm\u00eb m\u00ebnyrat p\u00ebr t'i zbuluar ata.
\n<\/a><\/noindex><\/p>\n

Mashtruesit e t\u00eb dh\u00ebnave jan\u00eb kudo<\/h2>\n

\nMashtruesit e t\u00eb dh\u00ebnave jan\u00eb aq t\u00eb mir\u00eb n\u00eb fshehjen e tyre, saq\u00eb ju mund t\u00eb jini nj\u00eb prej tyre<\/i>, duke mos e ditur fare. \u00cbsht\u00eb shum\u00eb e mundshme q\u00eb organizata juaj ka fshehur k\u00ebta mjesht\u00ebr p\u00ebr vite me radh\u00eb, por ka nj\u00eb lajm t\u00eb mir\u00eb: ata jan\u00eb t\u00eb leht\u00eb p\u00ebr t'u identifikuar n\u00ebse e dini \u00e7far\u00eb t\u00eb k\u00ebrkoni.
\nShenja e par\u00eb paralajm\u00ebruese \u00ebsht\u00eb moskuptimi se analiza dhe statistika jan\u00eb disciplina shum\u00eb t\u00eb ndryshme<\/strong>. M\u00eb tutje, do t'i shpjegohem k\u00ebt\u00eb.<\/p>\n

Disiplina t\u00eb ndryshme<\/h2>\n

\nStatistikat jan\u00eb t\u00eb trajnuar t\u00eb nxjerrin p\u00ebrfundime p\u00ebr at\u00eb q\u00eb del jasht\u00eb t\u00eb dh\u00ebnave t\u00eb tyre, analist\u00ebt jan\u00eb t\u00eb trajnuar t\u00eb studiojn\u00eb p\u00ebrmbajtjen e grupit t\u00eb t\u00eb dh\u00ebnave. Me fjal\u00eb t\u00eb tjera, analist\u00ebt nxjerrin p\u00ebrfundime mbi at\u00eb q\u00eb p\u00ebrmban informacioni i tyre, nd\u00ebrsa statistikat nxjerrin p\u00ebrfundime mbi at\u00eb q\u00eb mungon n\u00eb t\u00eb dh\u00ebnat. Analist\u00ebt ndihmojn\u00eb n\u00eb formulimin e pyetjeve t\u00eb mira (formulimin e hipotezave), nd\u00ebrsa statistikat ndihmojn\u00eb n\u00eb marrjen e p\u00ebrgjigjeve t\u00eb mira (verifikimin e hipotezave).<\/p>\n

Ekzistojn\u00eb gjithashtu role hibrid\u00eb t\u00eb \u00e7uditsh\u00ebm, kur nj\u00eb person p\u00ebrpiqet t\u00eb q\u00ebndroj\u00eb n\u00eb dy karrige... Pse jo? Parimi kryesor i shkenc\u00ebs s\u00eb t\u00eb dh\u00ebnave: n\u00ebse p\u00ebrballesh me pasiguri, nuk mund t\u00eb p\u00ebrdor\u00ebsh t\u00eb nj\u00ebjt\u00ebn<\/i> pik\u00eb t\u00eb dh\u00ebnash p\u00ebr hipoteza dhe verifikim. Kur t\u00eb dh\u00ebnat jan\u00eb t\u00eb kufizuara, pasiguria t\u00eb detyron t\u00eb zgjidh\u00ebsh midis statistikave ose analitik\u00ebs. Shpjegimi<\/i> k\u00ebtu<\/i><\/a><\/noindex>.<\/p>\n

Pa statistika do t\u00eb ngecni dhe nuk do t\u00eb kuptoni n\u00ebse vet\u00eb p\u00ebrfundimi i formuluar sapo \u00ebsht\u00eb i q\u00ebndruesh\u00ebm ndaj kritik\u00ebs, dhe pa analiz\u00eb l\u00ebvizni n\u00eb err\u00ebsir\u00eb, me pak shanse p\u00ebr t\u00eb p\u00ebrshkruar t\u00eb panjohur\u00ebn. Kjo \u00ebsht\u00eb nj\u00eb zgjedhje e v\u00ebshtir\u00eb.<\/p>\n

Shp\u00ebrthimi i nj\u00eb mashtruesi nga kjo situat\u00eb \u00ebsht\u00eb ta injorosh at\u00eb, e m\u00eb pas t\u00eb pretendoj\u00eb se \u00ebsht\u00eb i befasuar nga ajo q\u00eb papritur zbulohet. Logjika e kontrollit t\u00eb hipotezave statistike reduktohet n\u00eb pyetjen: a jan\u00eb t\u00eb dh\u00ebnat mjaft befasuese p\u00ebr t\u00eb ndryshuar mendimin ton\u00eb? Si mund t\u00eb jemi t\u00eb befasuar nga t\u00eb dh\u00ebnat n\u00ebse i kemi par\u00eb tashm\u00eb?<\/p>\n

Sa her\u00eb q\u00eb mashtruesit gjejn\u00eb nj\u00eb model, ata frym\u00ebzohen, pastaj e kontrollojn\u00eb t\u00eb nj\u00ebjtat t\u00eb dh\u00ebna<\/i> p\u00ebr t\u00eb t\u00eb nj\u00ebjtit model<\/i>, p\u00ebr t\u00eb publikuar nj\u00eb rezultat me nj\u00eb p-vler\u00eb legjitime ose dy, pran\u00eb teoris\u00eb s\u00eb tyre. Duke vepruar k\u00ebshtu, ata ju g\u00ebnjejn\u00eb (ndoshta edhe veten e tyre). Nj\u00eb p-vler\u00eb e till\u00eb nuk ka r\u00ebnd\u00ebsi n\u00ebse nuk q\u00ebndroni pas hipotez\u00ebs tuaj deri te<\/b> p\u00ebr m\u00ebnyr\u00ebn se si e keni par\u00eb t\u00eb dh\u00ebnat tuaja. Mashtruesit imitojn\u00eb veprimet e analist\u00ebve dhe statistik\u00ebve pa kuptuar arsyet. Si pasoj\u00eb, e gjith\u00eb fusha e shkenc\u00ebs t\u00eb dh\u00ebnave nxjerr nj\u00eb reputacion t\u00eb keq.<\/p>\n

Statistik\u00ebt e v\u00ebrtet\u00eb gjithmon\u00eb nxjerrin p\u00ebrfundimet e tyre<\/h2>\n

\nFal\u00eb reputacionit pothuajse mistik t\u00eb specialist\u00ebve n\u00eb statistik\u00eb, t\u00eb angazhuar me arsyetim rigoroz, sasia e informacionit t\u00eb rrem\u00eb n\u00eb Data Science \u00ebsht\u00eb rekordisht e lart\u00eb. \u00cbsht\u00eb leht\u00ebsisht p\u00ebr t'u mashtruar dhe p\u00ebr t'u kapur, ve\u00e7an\u00ebrisht n\u00ebse viktima e paq\u00ebndrueshme mendon se gjith\u00e7ka \u00ebsht\u00eb \u00e7\u00ebshtje e ekuacioneve dhe t\u00eb dh\u00ebnave. Nj\u00eb set t\u00eb dh\u00ebnash \u00ebsht\u00eb nj\u00eb set t\u00eb dh\u00ebnash, apo jo? Jo. Ka r\u00ebnd\u00ebsi si e p\u00ebrdorni at\u00eb.<\/p>\n

P\u00ebr fat t\u00eb mir\u00eb, ju nevojitet vet\u00ebm nj\u00eb sugjerim p\u00ebr t\u00eb kapur mashtruesit: ata \u00abrihapin Amerik\u00ebn pas ngjarjeve\u00bb. Rihapin fenomene q\u00eb, si\u00e7 din\u00eb ata tashm\u00eb, jan\u00eb t\u00eb pranishme n\u00eb t\u00eb dh\u00ebna. <\/p>\n

Ndryshe nga mashtruesit, analist\u00ebt e mir\u00eb nuk kan\u00eb paragjykime dhe kuptojn\u00eb se ide inspiruese mund t\u00eb ken\u00eb shum\u00eb shpjegime t\u00eb ndryshme. N\u00eb t\u00eb nj\u00ebjt\u00ebn koh\u00eb, statistik\u00ebt e mir\u00eb e p\u00ebrcaktojn\u00eb me kujdes p\u00ebrfundimin e tyre para se ta nxjerrin at\u00eb.<\/p>\n

Analist\u00ebt jan\u00eb t\u00eb liruar nga p\u00ebrgjegj\u00ebsia\u2026 p\u00ebrsa koh\u00eb ata nuk kalojn\u00eb p\u00ebrtej t\u00eb dh\u00ebnave t\u00eb tyre. N\u00ebse ata ndihen t\u00eb ndikuar t\u00eb deklarojn\u00eb di\u00e7ka q\u00eb nuk kan\u00eb par\u00eb, kjo \u00ebsht\u00eb nj\u00eb pun\u00eb krejt\u00ebsisht ndryshe. Ata duhet \"t\u00eb heqin k\u00ebpuc\u00ebt\" e analistit dhe \"t\u00eb veshin\" k\u00ebpuc\u00ebt e statistikani. N\u00eb fund t\u00eb fundit, pavar\u00ebsisht se \u00e7far\u00eb titulli zyrtar ka, nuk ka rregull q\u00eb ju ndalon t\u00eb studioni t\u00eb dyja profesione n\u00ebse d\u00ebshironi. Vet\u00ebm mos i ngat\u00ebrroni ato.<\/p>\n

N\u00ebse e keni mir\u00eb statistik\u00ebn, nuk do t\u00eb thot\u00eb q\u00eb e kuptoni mir\u00eb analitik\u00ebn, dhe e anasjelltas. N\u00ebse dikush p\u00ebrpiqet t'ju thot\u00eb t\u00eb kund\u00ebrt\u00ebn, \u00ebsht\u00eb e men\u00e7ur t\u00eb jeni t\u00eb kujdessh\u00ebm. N\u00ebse ky person ju informon se \u00ebsht\u00eb e lejueshme t\u00eb b\u00ebni p\u00ebrfundime statistike mbi t\u00eb dh\u00ebnat q\u00eb keni studiuar tashm\u00eb, kjo \u00ebsht\u00eb nj\u00eb arsye p\u00ebr t\u00eb qen\u00eb dyfish t\u00eb kujdessh\u00ebm.<\/p>\n

Shpjegime t\u00eb \u00e7uditshme<\/h2>\n

\nDuke v\u00ebzhguar mashtruesit e t\u00eb dh\u00ebnave n\u00eb natyr\u00eb, do t\u00eb vini re se ata i p\u00eblqejn\u00eb t\u00eb krijojn\u00eb histori fantastike p\u00ebr t\u00eb \"shpjeguar\" t\u00eb dh\u00ebnat e v\u00ebzhguara. Sa m\u00eb akademik\u00eb, aq m\u00eb mir\u00eb. Nuk ka r\u00ebnd\u00ebsi q\u00eb k\u00ebto histori jan\u00eb t\u00eb p\u00ebrshtatura pas t\u00eb fakt\u00ebs.<\/p>\n

Kur\u00eb b\u00ebjn\u00eb k\u00ebshtu fam\u00ebm\u00ebdhenjt\u00eb \u2014 lejoni q\u00eb t\u00eb mos kursej fjal\u00ebt \u2014 ata g\u00ebnjejn\u00eb. Asnj\u00eb sasi ekuacionesh apo koncepte t\u00eb bukura nuk kompenson at\u00eb q\u00eb ata ofruan si prov\u00eb t\u00eb zero p\u00ebr versionet e tyre. Mos u habitni se sa t\u00eb pazakonta jan\u00eb shpjegimet e tyre.<\/p>\n

Kjo \u00ebsht\u00eb nj\u00ebsoj si t\u00eb demonstrosh aft\u00ebsit\u00eb e tua \"ekstrasensoriale\", duke par\u00eb fillimisht kartat n\u00eb duar, dhe pastaj t\u00eb parashikosh se \u00e7far\u00eb mbani... at\u00eb q\u00eb mbani. Kjo \u00ebsht\u00eb nj\u00eb paragjykim k\u00ebndv\u00ebshtrimi retrospektiv, dhe profesionin e shkenc\u00ebtarit t\u00eb t\u00eb dh\u00ebnave e mbush me t\u00eb.<\/p>\n

\"Si<\/p>\n

Analist\u00ebt thon\u00eb: \"Sapo shkove me mbret\u00ebresh\u00ebn e kuqe.\" Statistikan\u00ebt thon\u00eb: \"E kam regjistruar hipotez\u00ebn time n\u00eb k\u00ebt\u00eb cop\u00eb letre p\u00ebrpara se t\u00eb fillonim. Le t\u00eb luajm\u00eb, t\u00eb shohim disa t\u00eb dh\u00ebna dhe t\u00eb shohim n\u00ebse kam t\u00eb drejt\u00eb.\" Fam\u00ebm\u00ebdhenjt\u00eb thon\u00eb: \"E dija se do t\u00eb shkoje me k\u00ebt\u00eb mbret\u00ebresh\u00ebn e kuqe, sepse...\" <\/p>\n

Shk\u00ebputja e t\u00eb dh\u00ebnave \u00ebsht\u00eb nj\u00eb zgjidhje e shpejt\u00eb p\u00ebr problemin q\u00eb \u00e7do person ka nevoj\u00eb.<\/p><\/blockquote>\n

\nKur t\u00eb dh\u00ebnat nuk jan\u00eb shum\u00eb, duhet t\u00eb zgjidhni mes statistik\u00ebs dhe analiz\u00ebs, por kur ka tep\u00ebr t\u00eb dh\u00ebna, ka nj\u00eb mund\u00ebsi t\u00eb shk\u00eblqyer p\u00ebr t\u00eb b\u00ebr\u00eb analiz\u00eb pa mashtrime. dhe<\/i> statistik\u00ebs. Keni mbrojtjen perfekte nga mashtruesit \u2014 kjo \u00ebsht\u00eb ndarja e t\u00eb dh\u00ebnave dhe, sipas mendimit tim, kjo \u00ebsht\u00eb ideja m\u00eb e fuqishme n\u00eb Data Science. <\/p>\n

P\u00ebr t'u mbrojtur nga mashtruesit, gjith\u00e7ka q\u00eb duhet t\u00eb b\u00ebni \u00ebsht\u00eb t\u00eb siguroheni q\u00eb t\u00eb ruhen disa t\u00eb dh\u00ebna testuese jasht\u00eb syve t\u00eb tyre kurioz\u00eb, dhe pastaj trajtoni gjith\u00e7ka tjet\u00ebr si analiz\u00eb. Kur hasni nj\u00eb teori q\u00eb rrezikoni t\u00eb pranoni, p\u00ebrdorni at\u00eb p\u00ebr t\u00eb vler\u00ebsuar situat\u00ebn dhe pastaj hapni t\u00eb dh\u00ebnat tuaja sekrete testuese p\u00ebr t\u00eb verifikuar se teoria nuk \u00ebsht\u00eb ndonj\u00eb gj\u00eb e \u00e7mendur. \u00cbsht\u00eb kaq e thjesht\u00eb!<\/p>\n

\"Si
\nSigurohuni q\u00eb askush t\u00eb mos lejohet t\u00eb shikoj\u00eb t\u00eb dh\u00ebnat testuese gjat\u00eb faz\u00ebs s\u00eb hulumtimit. P\u00ebr k\u00ebt\u00eb, q\u00ebndroni tek t\u00eb dh\u00ebnat hulumtuese. T\u00eb dh\u00ebnat testuese nuk duhet t\u00eb p\u00ebrdoren p\u00ebr analiz\u00eb.<\/p>\n

Kjo \u00ebsht\u00eb nj\u00eb hap i madh p\u00ebrpara krahasuar me at\u00eb q\u00eb njer\u00ebzit kan\u00eb m\u00ebsuar gjat\u00eb epok\u00ebs s\u00eb 't\u00eb dh\u00ebnave t\u00eb vogla', kur ju nevojitej t\u00eb shpjegoni se nga e dini at\u00eb q\u00eb dini, p\u00ebr t\u00eb bindur p\u00ebrfundimisht njer\u00ebzit se ju me t\u00eb v\u00ebrtet\u00eb dini di\u00e7ka.<\/p>\n

Aplikoni t\u00eb nj\u00ebjtat rregulla p\u00ebr ML\/AI<\/h2>\n

\nDisa mashtrues q\u00eb paraqiten si ekspert\u00eb n\u00eb fush\u00ebn e ML\/AI jan\u00eb gjithashtu t\u00eb leht\u00eb p\u00ebr tu zbuluar. Do t'i kapni ata ashtu si do t\u00eb kapnit \u00e7do inxhinier t\u00eb keq tjet\u00ebr: 'zgjidhjet', q\u00eb ata p\u00ebrpiqen t\u00eb ndihmojn\u00eb, d\u00ebshtojn\u00eb vazhdimisht. Nj\u00eb shenj\u00eb paralajm\u00ebruese e hershme \u00ebsht\u00eb mungesa e p\u00ebrvoj\u00ebs me gjuh\u00ebt dhe bibliotekat standarde t\u00eb industris\u00eb.<\/p>\n

Por \u00e7far\u00eb ndodh me njer\u00ebzit q\u00eb krijojn\u00eb sisteme q\u00eb duken funksionale? Si t\u00eb kuptoni se po ndodh di\u00e7ka t\u00eb dyshimt\u00eb? Rregulli i nj\u00ebjt\u00eb zbatohet! Mashtruesi \u00ebsht\u00eb nj\u00eb karakter i dyshimt\u00eb q\u00eb ju tregon sa mir\u00eb ka punuar modeli\u2026 me t\u00eb nj\u00ebjtat t\u00eb dh\u00ebna q\u00eb ata p\u00ebrdor\u00ebn p\u00ebr t\u00eb krijuar modelin.<\/p>\n

N\u00ebse keni krijuar nj\u00eb sistem absurdisht t\u00eb komplikuar t\u00eb m\u00ebsimit t\u00eb makinerive, si t\u00eb kuptoni se sa i mir\u00eb \u00ebsht\u00eb? Nuk do ta dini derisa t\u00eb tregoni se si funksionon me t\u00eb dh\u00ebna t\u00eb reja q\u00eb nuk i ka par\u00eb m\u00eb par\u00eb.<\/p>\n

Kur e keni par\u00eb t\u00eb dh\u00ebnat para parashikimit \u2014 dyshoj se \u00ebsht\u00eb para<\/i>deklarat\u00ebs.<\/p><\/blockquote>\n

\nKur keni mjaft t\u00eb dh\u00ebna p\u00ebr ndarje, nuk keni nevoj\u00eb t\u00eb referoheni n\u00eb bukurin\u00eb e formulave tuaja p\u00ebr t\u00eb justifikuar projektin (nj\u00eb zakon i vjet\u00ebr q\u00eb e shoh kudo, jo vet\u00ebm n\u00eb shkenc\u00eb). Mund t\u00eb thoni: \u00abE di q\u00eb funksionon, sepse mund t\u00eb marr\u00eb nj\u00eb grup t\u00eb dh\u00ebnash q\u00eb nuk e kam par\u00eb m\u00eb par\u00eb dhe t\u00eb parashikoj me sakt\u00ebsi se \u00e7far\u00eb do ndodh\u00eb... dhe do jem i sakt\u00eb. P\u00ebrs\u00ebri e p\u00ebrs\u00ebri\u00bb.<\/i><\/p>\n

Testimi i modelit tuaj\/teoris\u00eb me t\u00eb dh\u00ebna t\u00eb reja \u00ebsht\u00eb baza m\u00eb e mir\u00eb p\u00ebr besim.<\/p><\/blockquote>\n

\nNuk i toleroj mashtruesit e t\u00eb dh\u00ebnave. S'm\u00eb intereson n\u00ebse opinioni juaj mb\u00ebshtetet n\u00eb karakteristika t\u00eb ndryshme. Bukuria e shpjegimeve tuaja nuk m\u00eb impresionon. M\u00eb tregoni se teoria\/juaj model funksionon (dhe vazhdon t\u00eb funskionoj\u00eb) me nj\u00eb s\u00ebr\u00eb t\u00eb dh\u00ebnash t\u00eb reja, q\u00eb nuk keni par\u00eb ndonj\u00ebher\u00eb m\u00eb par\u00eb. Kjo \u00ebsht\u00eb v\u00ebrtet provimi i q\u00ebndrueshm\u00ebris\u00eb s\u00eb opinionit tuaj.<\/p>\n

Dhe k\u00ebrkoni ndihm\u00ebn e profesionist\u00ebve n\u00eb fush\u00ebn e Data Science<\/h2>\n

\nN\u00ebse doni t\u00eb respektoheni nga t\u00eb gjith\u00eb ata q\u00eb e kuptojn\u00eb k\u00ebt\u00eb humor, ndaloni s\u00eb fshehuri pas ekuacioneve t\u00eb \u00e7uditshme p\u00ebr t\u00eb mb\u00ebshtetur paragjykimet personale. Tregoni se \u00e7far\u00eb keni. N\u00ebse d\u00ebshironi q\u00eb ata q\u00eb \"kan\u00eb kuptuar\" ta shohin teorin\u00eb \/ modelin tuaj si di\u00e7ka m\u00eb shum\u00eb se nj\u00eb poezi frym\u00ebzuese, keni guximin t\u00eb organizoni nj\u00eb shfaqje grandioze p\u00ebr sa mir\u00eb funksionon n\u00eb nj\u00eb grup t\u00eb ri t\u00eb dh\u00ebnash\u2026 para d\u00ebshmitar\u00ebve!<\/p>\n

K\u00ebrkesa p\u00ebr menaxher\u00ebt<\/h2>\n

\nRefuzoni t\u00eb merrni seriozisht \u00e7do \"ide\" p\u00ebr t\u00eb dh\u00ebnat derisa t\u00eb provohen n\u00eb t\u00eb dh\u00ebna t\u00eb reja.<\/i> Nuk doni t\u00eb b\u00ebni p\u00ebrpjekje? Q\u00ebndroni te analitika, por mos u mb\u00ebshtetni te k\u00ebto ide \u2014 ato jan\u00eb t\u00eb pasigurta dhe nuk jan\u00eb verifikuar p\u00ebr besueshm\u00ebri. P\u00ebr m\u00eb tep\u00ebr, kur nj\u00eb organizat\u00eb ka t\u00eb dh\u00ebna n\u00eb boll\u00ebk, nuk ka asnj\u00eb defekt n\u00eb b\u00ebrjen e ndarjes nj\u00eb themel n\u00eb shkenc\u00eb dhe ta mbani at\u00eb n\u00eb nivelin e infrastruktur\u00ebs duke kontrolluar aksesin n\u00eb t\u00eb dh\u00ebnat testuese p\u00ebr statistikat. Kjo \u00ebsht\u00eb nj\u00eb m\u00ebnyr\u00eb e shk\u00eblqyer p\u00ebr t\u00eb parandaluar p\u00ebrpjekjet p\u00ebr t'ju mashtruar!<\/p>\n

N\u00ebse d\u00ebshironi t\u00eb shihni m\u00eb shum\u00eb shembuj t\u00eb mashtruesve q\u00eb planifikojn\u00eb di\u00e7ka t\u00eb keqe \u2014 ky \u00ebsht\u00eb nj\u00eb thurje e shk\u00eblqyer n\u00eb Twitter<\/a><\/noindex>.<\/p>\n

P\u00ebrfundimet<\/h2>\n

\nKur ka shum\u00eb pak t\u00eb dh\u00ebna p\u00ebr ndarje, vet\u00ebm nj\u00eb mashtrues p\u00ebrpiqet t\u00eb ndjek\u00eb me rigorozitet frym\u00ebzimin, duke zbuluar Amerik\u00ebn retrospektivisht, duke e rikthyer matematikisht fenomenin q\u00eb tashm\u00eb \u00ebsht\u00eb i njohur n\u00eb t\u00eb dh\u00ebna, dhe duke e quajtur mrekullin\u00eb statistike t\u00eb r\u00ebnd\u00ebsishme. Kjo i ndan ata nga analisti pa paragjykime q\u00eb merret me frym\u00ebzimin dhe statistiku i kujdessh\u00ebm q\u00eb ofron prova kur parashikon.<\/p>\n

Kur ka shum\u00eb t\u00eb dh\u00ebna, zhvilloni zakonin e ndarjes s\u00eb t\u00eb dh\u00ebnave, k\u00ebshtu do keni m\u00eb t\u00eb mirat e t\u00eb dyja bot\u00ebve! Sigurohuni q\u00eb t\u00eb b\u00ebni analitik\u00eb dhe statistika ndaras p\u00ebr n\u00ebngrupet e ndryshme t\u00eb grumbullimit fillestar t\u00eb t\u00eb dh\u00ebnave.<\/p>\n