DA0-002 Data Analysis Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✓

  To find the point where the rate of decrease in WCSS slows down

  Why this is correct

  The elbow method identifies the k where WCSS reduction sharply decelerates, marking diminishing returns from adding clusters. Plotting WCSS against k, the inflection point balances model complexity against fit, so the marketing team selects the smallest k beyond which further segmentation yields negligible within-cluster variance improvement.

• ✗

  To identify the value of k that minimizes WCSS

  Why it's wrong here

  Minimising WCSS alone always favours the largest k, so the elbow method instead finds the point where added clusters yield diminishing WCSS reduction. Minimisation is tempting because WCSS measures cluster compactness, but it would select k equal to the number of data points.

• ✗

  To determine the initial centroids for the algorithm

  Why it's wrong here

  Plotting WCSS against k identifies the elbow where adding clusters yields diminishing variance reduction, which is unrelated to seeding initial centroids. It is tempting because centroid initialisation genuinely affects K-means convergence, and k-means++ is the correct technique when the goal is choosing starting points rather than the cluster count.

• ✗

  To ensure all clusters have equal size

  Why it's wrong here

  The elbow method plots WCSS against k to find where adding clusters stops yielding meaningful variance reduction; it says nothing about cluster cardinality, since K-means naturally produces unevenly sized clusters. Equal-size segmentation would instead call for constrained clustering such as balanced k-means, so this option misreads the plot's purpose entirely.