Nj\u00eb profesor Universiteti n\u00eb Kanada ka gjetur nj\u00eb metod\u00eb inovative p\u00ebr t\u00eb zbuluar plagjiatur\u00ebn e krijuar nga AI, duke paraqitur k\u00ebshtu nj\u00eb mjet t\u00eb ri n\u00eb betej\u00ebn kund\u00ebr mashtrimit akademik. Kjo teknik\u00eb p\u00ebrqendrohet kryesisht n\u00eb dallimin e autenticitetit t\u00eb citimeve t\u00eb p\u00ebrdorura n\u00eb artikujt akademik\u00eb.<\/p>\n\n\n\n
Profesor Terence Day, nj\u00eb profesor i gjeografis\u00eb fizike n\u00eb Kolegjin Okanagan, British Columbia, e gjeti veten t\u00eb mbytur nga punimet e krijuara nga AI, t\u00eb cilat dor\u00ebzoheshin nga student\u00ebt si detyra autentike. Sipas tij, punimet e krijuara nga AI mund t\u00eb verifikohen duke kontrolluar referencat e vendosura n\u00eb fund t\u00eb punimeve, t\u00eb cilat shpeshher\u00eb nuk jan\u00eb t\u00eb sakta. “T\u00eb dallosh k\u00ebto referenca t\u00eb rreme \u00ebsht\u00eb mjaft e thjesht\u00eb,” tha Day. “Nj\u00eb verifikim i thjesht\u00eb n\u00eb internet mund t\u00eb zbuloj\u00eb vlefshm\u00ebrin\u00eb e tyre.”<\/p>\n\n\n\n
N\u00eb propozimin e tij t\u00eb fundit, Day thekson se student\u00ebve duhet t’u k\u00ebrkohet t\u00eb shtojn\u00eb nj\u00eb identifikues digjital t\u00eb objektit (DOI) p\u00ebr \u00e7do citim n\u00eb pun\u00ebt e tyre akademike. \u201cDuke klikuar n\u00eb k\u00ebto link-e, ne mund t\u00eb konstatojm\u00eb shpejt n\u00ebse punimi i cituar ekziston apo jo\u201d, shpjegoi Day.<\/p>\n\n\n\n
Kjo procedur\u00eb u p\u00ebrshkrua p\u00ebr her\u00eb t\u00eb par\u00eb n\u00eb detaje n\u00eb punimin k\u00ebrkimor t\u00eb rishikuar nga koleg\u00ebt e Day, punim i botuar n\u00eb The Professional Geographer. Kurioziteti i Day p\u00ebr citimet e krijuara nga AI lindi gjat\u00eb eksperimenteve t\u00eb tij me ChatGPT, nj\u00eb model AI, i cili prodhoi citime q\u00eb n\u00eb dukje ishin legjitime p\u00ebr pyetjet e tij n\u00eb fush\u00ebn e gjeografis\u00eb.<\/p>\n\n\n\n
Megjithat\u00eb, n\u00eb nj\u00eb analiz\u00eb m\u00eb t\u00eb detajuar, Day zbuloi se k\u00ebto referenca ishin t\u00eb rreme. “Pas ekzaminimit t\u00eb citimeve t\u00eb krijuara nga ChatGPT, kuptova se t\u00eb gjitha ishin krejt\u00ebsisht t\u00eb rreme. Ato nuk ekzistonin n\u00eb revistat apo v\u00ebllimet e specifikuara, dhe as nuk u gjet\u00ebn n\u00eb Google Scholar,” tha ai. “Asnj\u00eb nga referencat e shumta q\u00eb un\u00eb kontrollova nuk doli t\u00eb ishte e sakt\u00eb apo ekzistuese.” Day e shpjegon k\u00ebt\u00eb mosp\u00ebrputhje me parashikimet algoritmike t\u00eb ChatGPT, t\u00eb cilat kusht\u00ebzohen nga trajnimi i AI n\u00eb nj\u00eb fush\u00eb t\u00eb caktuar. K\u00ebto citime, sipas tij, ishin kryesisht t\u00eb fabrikuara.<\/p>\n\n\n\n
N\u00eb gar\u00ebn p\u00ebr t\u00eb zhvilluar mjete p\u00ebr zbulimin e plagjiatur\u00ebs s\u00eb krijuar nga AI, metoda e Day ofron nj\u00eb alternativ\u00eb q\u00eb ai beson se \u00ebsht\u00eb m\u00eb e besueshme. \u201cNd\u00ebrsa software p\u00ebr zbulimin e plagjiatur\u00ebs jep vet\u00ebm nj\u00eb probabilitet t\u00eb plagjiatur\u00ebs, ai nuk mund t\u00eb p\u00ebrcaktoj\u00eb p\u00ebrfundimisht n\u00ebse puna \u00ebsht\u00eb krijuar nga nj\u00eb njeri, apo nga AI\u201d, tha Day.<\/p>\n\n\n\n
Qasja e profesorit, e bazuar n\u00eb kontrollin e autenticitetit t\u00eb citimeve, ofron nj\u00eb tregues t\u00eb qart\u00eb t\u00eb matjes s\u00eb ndershm\u00ebris\u00eb akademike. “Forca e k\u00ebsaj metode q\u00ebndron n\u00eb thjesht\u00ebsin\u00eb dhe rezultatet e saj. N\u00ebse referenca e cituar ekziston, ka t\u00eb ngjar\u00eb q\u00eb puna t\u00eb jet\u00eb e v\u00ebrtet\u00eb,” shpjegoi Day.<\/p>\n\n\n\n
Paralajm\u00ebrimi i tij ndaj student\u00ebve ishte i prer\u00eb: “N\u00ebse punimi i cituar nuk ekziston, ju kapeni n\u00eb flagranc\u00eb”. Kjo metod\u00eb e re mund t\u00eb sjell\u00eb nj\u00eb ndryshim n\u00eb kontrollet e integritetit akademik, duke siguruar nj\u00eb t\u00eb ardhme ku keqp\u00ebrdorimi i mundsh\u00ebm i inteligjenc\u00ebs artificiale n\u00eb plagjiatur\u00ebn akademike mund t\u00eb kufizohet n\u00eb m\u00ebnyr\u00eb efektive.<\/p>\n","protected":false},"excerpt":{"rendered":"
Nj\u00eb profesor Universiteti n\u00eb Kanada ka gjetur nj\u00eb metod\u00eb inovative p\u00ebr t\u00eb zbuluar plagjiatur\u00ebn e krijuar nga AI, duke paraqitur k\u00ebshtu nj\u00eb mjet t\u00eb ri n\u00eb betej\u00ebn kund\u00ebr mashtrimit akademik. Kjo teknik\u00eb p\u00ebrqendrohet kryesisht n\u00eb dallimin e autenticitetit t\u00eb citimeve t\u00eb p\u00ebrdorura n\u00eb artikujt akademik\u00eb. Profesor Terence Day, nj\u00eb profesor i gjeografis\u00eb fizike n\u00eb Kolegjin<\/p>\n","protected":false},"author":4,"featured_media":13567,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[38],"tags":[40,41],"class_list":{"0":"post-13559","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-tech","8":"tag-featured","9":"tag-picks"},"yoast_head":"\n