Kedis ku xidhan hagaha · Dhexdhexaad Heerka

ROC Curves iyo AUC Kedis

Hubi waxa ROC-AUC cabbiro, sida ay ula xidhiidho darajooyinka iyo heerarka, iyo sababta isku dheelitirnaanta ama isu dheelitir la'aanta ay ugu baahan yihiin falanqayn gooni ah.

Wadooyinka haga ee la xidhiidhaRoc Curves iyo Auc
Su'aal 1 ee 8

Waa maxay labada qiyaasood ee ROC uu midba midka kale u maleego?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Waa maxay labada qiyaasood ee ROC uu midba midka kale u maleego?
    • Qiimaha Toosan ee Runta ah vs. Qiimaha Wanaagsan ee Beenta
    • Saxnimada vs. Xusuusin
    • Saxnimada vs. Xadka
    • Dareenka vs. Saxnaanta
  2. Waa maxay qiimaha AUC ee u dhigma kala-soociye aan qabanin wax ka fiican male-awaal aan toos ahayn?
    • 0.0
    • 0.5
    • 0.75
    • 1.0
  3. AUC waxaa loo tarjumi karaa inay tahay ixtimaalka in moodelku:
    • Si sax ah u kala saara muunad si aan kala sooc lahayn loo doortay
    • Wuxuu gaaraa in ka badan 90% sax
    • Dhibcaha togan ee random ka sareeyo taban random
    • Leh eber xaqiiqooyin been ah
  4. Waa maxay sababta AUC loogu sifeeyay 'madax-bannaanida xadka'?
    • Waxa kaliya oo ay ka shaqeysaa 0.5 marin
    • Waxay iska indha tirtaa fasalka togan gabi ahaanba
    • Waxay u baahan tahay ixtimaalka la saxay
    • Waxay isku daraysaa waxqabadka dhammaan heerarka suurtagalka ah
  5. Xogta aadka u dheelitiran, qalooca ayaa inta badan lagula taliyaa inay garab socoto ama halkii ROC sababtoo ah ROC waxay u eegi kartaa rajo xad dhaaf ah?
    • Qallooca dib u soo celinta
    • qalooca waxbarashada
    • qalooca calibration
    • Qallooca luminta
  6. Bar ku taal geeska bidix ee sare ee booska ROC waxay u taagan tahay:
    • Waxyaalo been abuur ah oo sarreeya iyo wanaag run ah oo hooseeya
    • Kala saarid qumman ama u dhow
    • Waxqabadka random
    • Qaab khaldan
  7. Why can a model have a high ROC-AUC but poorly calibrated probability estimates?
    • AUC directly measures the distance between every predicted probability and its true label
    • Calibration is guaranteed whenever the ROC curve is above the diagonal
    • AUC depends on score ranking, so a monotonic score transformation can preserve ranking while changing probability values
    • AUC uses only one fixed classification threshold
  8. What does partial AUC let an evaluator emphasize?
    • Only the model's training loss
    • Probability calibration across all score values
    • The number of labels in a multiclass model
    • Performance in a selected false-positive-rate region, such as the low-FPR range