Kedis ku xidhan hagaha · Dhexdhexaad Heerka

K-Waxaa loola jeedaa Kediska Kooxaynta

Ku celceli shaqada K-Means, cusboonaysiinta centroid, firfircooni la'aanta, bilaabista, iyo qaabka qaab-kooxeedka oo wata tusaalooyin kooxeed la taaban karo.

Wadooyinka haga ee la xidhiidhaK Waxaa loola jeedaa Koox-kooxeed
Su'aal 1 ee 8

Maxaa 'K' ee K-Means loola jeedaa?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Maxaa 'K' ee K-Means loola jeedaa?
    • Tirada kutlada aad horay u doorato
    • Tirada dhibcaha xogta
    • Tirada celceliska tababarka
    • Tirada sifooyinka ku jira xogta
  2. Maxay yihiin labada tallaabo ee soo noqnoqda ee loop-ka-Means?
    • Kala qaybso rucubyada, ka dibna ku dar rucubyada
    • U qoondee dhibcaha centroid-ka kuugu dhow, ka dibna u wareeji centroids celceliska
    • Calaamadee xogta, ka dibna tababar kala saar
    • Kala sooc xogta, ka dibna ka saar kuwa ka baxsan
  3. Waa maxay tirada uu K-Means isku dayo inuu yareeyo?
    • Tirada kooxaha
    • Masaafada u dhaxaysa centroids
    • Masaafada labajibaaran ee kutlada gudahooda (inertia)
    • Tirada sifooyinka
  4. Waa maxay sababta K-Means loogu yeedhay algorithm 'aan la ilaalin'?
    • It uses predefined class labels to learn a boundary
    • It requires one label per initial centroid
    • It estimates labels only after supervised training
    • It groups inputs without target class labels
  5. Waa maxay 'habka suxulka' ee inta badan loo isticmaalo?
    • Dar-dargelinta tababarka
    • Doorashada qiimaha wanaagsan ee K
    • Bilaabidda centroids
    • Ka saarida dhibco nuqul ah
  6. Maxay K-Means ku siin kartaa natiijooyin kala duwan orodyo kala duwan?
    • K automatically changes on every run
    • Euclidean distance uses a random definition
    • Results depend on initial centroid positions
    • Convergence always produces the same global optimum
  7. A K-Means cluster contains the one-dimensional values 2, 4, and 9. Where does its centroid move in the update step?
    • 5, the mean of the assigned values
    • 4, the middle observed value
    • 9, the largest assigned value
    • 15, the sum of the assigned values
  8. A dataset contains two long, curved groups. Which K-Means limitation should the analyst consider?
    • The groups must have labels before clustering
    • K-Means always discovers curved boundaries exactly
    • Changing K guarantees the intended grouping
    • Euclidean distance to centroids can poorly represent elongated or non-spherical groups