Maswali yanayounganishwa na mwongozo · Kati Kiwango

ROC Curves na Maswali ya AUC

Angalia ni hatua zipi za ROC-AUC, jinsi inavyohusiana na nafasi na vizingiti, na kwa nini urekebishaji au usawa unahitaji uchanganuzi tofauti.

Njia za mwongozo zinazohusianaRoc Curves Na Auc
Swali 1 ya 8

Je, ROC inapanga njama gani dhidi ya kila mmoja?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Je, ROC inapanga njama gani dhidi ya kila mmoja?
    • Kiwango Chanya cha Kweli dhidi ya Kiwango Chanya cha Uongo
    • Usahihi dhidi ya Kumbuka
    • Usahihi dhidi ya Kizingiti
    • Unyeti dhidi ya Usahihi
  2. Ni thamani gani ya AUC inalingana na kiainishaji ambacho hakifanyi kazi bora kuliko kubahatisha bila mpangilio?
    • 0.0
    • 0.5
    • 0.75
    • 1.0
  3. AUC inaweza kufasiriwa kama uwezekano kwamba mfano:
    • Huainisha kwa usahihi sampuli iliyochaguliwa nasibu
    • Inafikia usahihi wa zaidi ya 90%.
    • Alama chanya nasibu juu kuliko hasi nasibu
    • Haina chanya sifuri za uwongo
  4. Kwa nini AUC inaelezewa kama 'kizingiti-huru'?
    • Inafanya kazi tu kwa kizingiti cha 0.5
    • Inapuuza darasa chanya kabisa
    • Inahitaji uwezekano uliorekebishwa
    • Inaunganisha utendaji katika viwango vyote vinavyowezekana
  5. Kwa mkusanyiko wa data usio na usawa, ni curve ipi inayopendekezwa mara nyingi kando au badala ya ROC kwa sababu ROC inaweza kuonekana kuwa na matumaini kupita kiasi?
    • Precision-Recall Curve
    • Curve ya kujifunza
    • Curve ya urekebishaji
    • Curve ya kupoteza
  6. Sehemu katika kona ya juu kushoto ya nafasi ya ROC inawakilisha:
    • Chanya za juu za uwongo na chanya za chini za kweli
    • Utengano kamili au karibu kabisa
    • Utendaji bila mpangilio
    • Mfano usiofaa
  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