Ukuhlolwa kwangaphandle kwedatha yethu yokwenziwa ochwepheshe bedatha bakwa-SAS
Idatha yethu yokwenziwa ithi ihlolwe futhi kuvunyelwe ngochwepheshe bedatha ye SAS
Ukwethulwa kokuhlolwa kwangaphandle kwedatha yethu yokwenziwa ochwepheshe bedatha be-SAS
Senzeni?
Idatha ye-synthetic ekhiqizwa yi-Syntho iyahlolwa, iqinisekiswe futhi ivunyelwe ngokombono wangaphandle nomgomo wochwepheshe bedatha ye-SAS.
Kungani idatha yethu yokwenziwa ihlolwa ngaphandle ngochwepheshe bedatha bakwa-SAS?
Nakuba i-Syntho iziqhenya ngokunikeza abasebenzisi bayo umbiko wokuqinisekisa ikhwalithi ethuthukisiwe, futhi siyakuqonda ukubaluleka kokuba nokuhlola kwangaphandle nokunenhloso kwedatha yethu yokwenziwa evela kubaholi bemboni. Yingakho sibambisana ne-SAS, umholi wezibalo, ukuze sihlole idatha yethu yokwenziwa.
I-SAS yenza ukuhlola okuphelele okuhlukahlukene kokunemba kwedatha, ukuvikelwa kobumfihlo, kanye nokusebenziseka kwedatha yokwenziwa ekhiqizwe yi-Syntho's AI uma kuqhathaniswa nedatha yokuqala. Njengokuphetha, i-SAS yahlola futhi yagunyaza idatha yokwenziwa ye-Syntho njengenembile, ivikelekile, futhi iyasebenziseka uma iqhathaniswa nedatha yokuqala.
Yini eyenziwa yi-SAS phakathi nalokhu kuhlola?
Sisebenzise idatha ye-telecom esetshenziselwa ukubikezela "i-churn" njengedatha eqondiwe. Umgomo wokuhlola kwakuwukusebenzisa idatha yokwenziwa ukuqeqesha amamodeli ahlukahlukene wokubikezela i-churn nokuhlola ukusebenza kwemodeli ngayinye. Njengoba ukubikezela kwe-churn kuwumsebenzi wokuhlukanisa, i-SAS ikhethe amamodeli okuhlukanisa adumile ukwenza izibikezelo, okuhlanganisa:
- Ihlathi elingahleliwe
- Ukuthuthukisa i-gradient
- Ukuhlehla kwezinto
- Inethiwekhi ye-Neural
Ngaphambi kokukhiqiza idatha yokwenziwa, i-SAS ihlukanisa ngokungahleliwe idathasethi yezokuxhumana ibe yisethi yesitimela (yokuqeqesha amamodeli) kanye nesethi yokubamba (ukuthola amamodeli). Ukuba nesethi ehlukile yokubamba amaphuzu kuvumela ukuhlola okungachemile kokuthi imodeli yokuhlukanisa ingenza kahle kanjani uma isetshenziswa kudatha entsha.
Isebenzisa isethi yesitimela njengokufakwayo, i-Syntho yasebenzisa i-Syntho Engine yayo ukuze ikhiqize idathasethi yokwenziwa. Ngokulinganisa, i-SAS iphinde yadala inguqulo engaziwa yesethi yesitimela ngemva kokusebenzisa amasu ahlukahlukene okwenza ungaziwa ukuze ufinyelele umkhawulo othile (wokungaziwa). Izinyathelo zangaphambili ziholele kumasethi edatha amane:
- Isethi yedatha yesitimela (okungukuthi, isethi yedatha yoqobo kukhishwe idathasethi yokubamba)
- Isethi yedatha yokubamba (okungukuthi isethi engaphansi yedathasethi yasekuqaleni)
- Isethi yedatha engaziwa (idatha engaziwa yedathasethi yesitimela, isethi yedatha yoqobo kukhishwe isethi yedatha yokubamba)
- Idathasethi yokwenziwa (idatha ehlanganisiwe yedathasethi yesitimela, isethi yedatha yoqobo kukhishwe isethi yedatha yokubamba)
Amasethi edatha 1, 3 kanye no-4 asetshenziselwa ukuqeqesha imodeli yokuhlukanisa ngayinye, okuholele kumamodeli aqeqeshiwe ayi-12 (3 x 4). I-SAS yabe isisebenzisa idathasethi ye-holdout ukukala ukunemba kwemodeli ngayinye ekubikezelweni kokuxokozela kwamakhasimende.
I-SAS yenza ukuhlola okuphelele okuhlukahlukene kokunemba kwedatha, ukuvikelwa kobumfihlo, kanye nokusebenziseka kwedatha yokwenziwa ekhiqizwe yi-Syntho's AI uma kuqhathaniswa nedatha yokuqala. Njengokuphetha, i-SAS yahlola futhi yagunyaza idatha yokwenziwa ye-Syntho njengenembile, ivikelekile, futhi iyasebenziseka uma iqhathaniswa nedatha yokuqala.
Ingabe unemibuzo?
Khuluma nomunye wochwepheshe bethu
Imiphumela yokuqala yokuhlolwa kwedatha yi-SAS
Amamodeli aqeqeshwe kudatha yokwenziwa athola amaphuzu afana kakhulu uma kuqhathaniswa namamodeli aqeqeshwe kudatha yangempela
Idatha yokwenziwa evela ku-Syntho ayibambeli nje kuphela amaphethini ayisisekelo, futhi ithwebula amaphethini ezibalo 'afihliwe' ajulile adingekayo emisebenzini yezibalo ezithuthukile. Lokhu kwakamuva kuboniswa eshadini lebha, okubonisa ukuthi ukunemba kwamamodeli aqeqeshwe kudatha yokwenziwa uma kuqhathaniswa namamodeli aqeqeshwe kudatha yoqobo ayafana. Ngakho-ke, idatha yokwenziwa ingasetshenziselwa ukuqeqeshwa kwangempela kwamamodeli. Okokufaka nokubaluleka okuguquguqukayo okukhethwe ama-algorithms kudatha yokwenziwa uma kuqhathaniswa nedatha yoqobo bekufana kakhulu. Ngakho-ke, kuphetha ngokuthi inqubo yokumodela ingenziwa kudatha yokwenziwa, njengenye indlela yokusebenzisa idatha ebucayi yangempela.
Kungani amamodeli aqeqeshelwe imiphumela yedatha engaziwa eba mabi kakhulu?
Izindlela zakudala zokungaziwa zifana nokuthi zikhohlisa idatha yoqobo ukuze zivimbele ukulandelela abantu ngabanye. Bakhohlisa idatha futhi ngaleyo ndlela bacekele phansi idatha kunqubo. Lapho ugcina ungadaluli igama lakho, idatha yakho ivikelwa kangcono, kodwa futhi idatha yakho yonakala kakhulu. Lokhu kulimaza kakhulu i-AI nemisebenzi yokumodela lapho “amandla okubikezela” ebalulekile, ngoba idatha yekhwalithi embi izophumela ekuqondeni okubi okuvela kumodeli ye-AI. I-SAS ikhombise lokhu, ngendawo engaphansi kwejika (AUC*) esondele ku-0.5, okubonisa ukuthi amamodeli aqeqeshelwe idatha engaziwa enza kabi kakhulu.
Imiphumela eyengeziwe yokuhlolwa kwedatha yokwenziwa yi-SAS
Imiphumela eyengeziwe yokuhlolwa kwedatha yokwenziwa yi-SAS
Ukuhlobana nobudlelwano phakathi kokuguquguqukayo kulondolozwe ngokunembile kudatha yokwenziwa.
I-Area Under the Curve (AUC), i-metric yokusebenza kwemodeli yokulinganisa, ihlale injalo.
Ngaphezu kwalokho, ukubaluleka okuguquguqukayo, okubonisa amandla okubikezela okuguquguqukayo kumodeli, kuhlale kuqinile lapho kuqhathaniswa idatha yokwenziwa nedathasethi yoqobo.
Ngokusekelwe kulokhu okuphawulwe yi-SAS nokusebenzisa i-SAS Viya, singaphetha ngokuzethemba ukuthi idatha yokwenziwa ekhiqizwe i-Syntho Engine ilingana ngempela nedatha yangempela ngokwekhwalithi. Lokhu kuqinisekisa ukusetshenziswa kwedatha yokwenziwa yokuthuthukiswa kwemodeli, kuvula indlela yokuhlaziya okuthuthukile ngedatha yokwenziwa.
Iziphetho ngochwepheshe bedatha be-SAS
- Amamodeli aqeqeshwe kudatha yokwenziwa uma kuqhathaniswa namamodeli aqeqeshwe kudatha yoqobo abonisa ukusebenza okufanayo kakhulu
- Amamodeli aqeqeshelwe idatha engaziwa 'anezindlela zakudala zokungaziwa' abonisa ukusebenza okuphansi uma kuqhathaniswa namamodeli aqeqeshwe kudatha yoqobo noma idatha yokwenziwa
- Ukwenziwa kwedatha yokwenziwa kulula futhi kuyashesha ngoba inqubo isebenza ngokufana ncamashi nedathasethi ngayinye kanye nohlobo lwedatha ngayinye
Izihloko eziyisethenjwa
- Ukuhlolwa kochwepheshe bedatha be-SAS: https://blogs.sas.com/content/hiddeninsights/2022/07/07/ai-generated-synthetic-data-easy-and-fast-access-to-high-quality-data/
- U-Syntho owine i-SAS global hackathon: https://www.linkedin.com/feed/update/urn:li:activity:7070047376249376769/
- Imiphumela yocwaningo lwezempilo: https://communities.sas.com/t5/SAS-Hacker-s-Hub/AI-Generated-Synthetic-Data-in-Healthcare/ta-p/863407
Londoloza umhlahlandlela wakho wedatha wokwenziwa manje!
- Iyini idatha yokwenziwa?
- Kungani izinhlangano ziyisebenzisa?
- Inani elengeza amakesi eklayenti edatha yokwenziwa
- Ungaqala kanjani
