CompTIA DY0-001 - CompTIA DataX Exam

→ Scatter plots are used to examine relationships and trends between two numeric variables. They are especially effective at revealing linear (or nonlinear) patterns, clusters, and outliers.

Why the other options are incorrect:

A: Violin plots show distribution and density, not relationships.

B: Box plots compare distributions across groups but don’t reveal linearity.

D: Q-Q plots test normality, not variable relationships.

Official References:

CompTIA DataX (DY0-001) Official Study Guide – Section 1.2:“Scatter plots are commonly used during EDA to identify correlations, linearity, and outliers between two continuous variables.”

Data Science Fundamentals, Chapter 2 – EDA Techniques:“To assess linear trends and relationships, scatter plots provide a direct visual assessment between variables.”

→ In the context of Naive Bayes classifiers, the "naive" assumption refers to the conditional independence of features given the class label. That is, the model assumes each feature contributes independently to the probability of the output class, which simplifies the computation of probabilities.

Why the other options are incorrect:

A: Normal distribution is often assumed for continuous variables, but it's not the naive assumption in Bayes' rule.

C: Uniform distribution refers to equal probability across outcomes, not used here.

D: Homoskedasticity is related to constant variance in regression, not Bayesian classification.

Official References:

CompTIA DataX (DY0-001) Study Guide – Section 4.1:“Naive Bayes assumes all features are conditionally independent given the target class, which allows for efficient computation.”

—

→ A Sankey diagram is ideal for illustrating flow-based relationships, such as how different units or sources contribute to a total. It's especially effective for showing proportions, hierarchy, and decomposition — such as revenue contribution by business units.

Why the other options are incorrect:

A: Box plots show distributions and spread — not contributions or breakdowns.

C: Scatter plot matrix explores relationships between numeric variables, not part-to-whole relationships.

D: Residual charts are diagnostic tools for regression — not for revenue visualization.

Official References:

CompTIA DataX (DY0-001) Official Study Guide – Section 5.5:“Sankey diagrams are useful for visualizing contributions, flows, and proportional allocations across categories.”

Data Visualization Best Practices, Chapter 7:“Sankey charts are preferred when tracking contributions from multiple inputs to a unified output.”

→ The graph represents a classic “elbow curve,” which is often used in clustering (e.g., k-means) to help determine the optimal number of clusters. The point where the curve starts to level off (the “elbow”) reflects the best trade-off between model accuracy and processing efficiency.

In this graph, the elbow visually occurs around k = 10. Beyond that, the processing time continues to decrease, but the marginal gain in clustering quality (or drop in processing time) diminishes.

Why the other options are incorrect:

A: k = 2 underfits the data — too few clusters.

C & D: k = 15 or 20 provides minimal additional benefit in processing but may overcomplicate the model.

Official References:

CompTIA DataX (DY0-001) Study Guide – Section 4.2:“The elbow method identifies the optimal number of clusters where the rate of improvement drops significantly.”

—

→ Pooling layers are used in Convolutional Neural Networks (CNNs) to reduce the spatial dimensions (width and height) of the feature maps. This helps in downsampling, reducing computational complexity, and controlling overfitting by summarizing the features (e.g., max pooling or average pooling).

Why the other options are incorrect:

B: Input layers receive raw data and do not perform downsampling.

C: Output layers generate the final prediction.

D: Hidden layers process data but do not specifically perform downsampling unless designed to do so (e.g., convolutional or pooling sublayers).

Official References:

CompTIA DataX (DY0-001) Study Guide – Section 4.3:“Pooling layers are used to downsample feature maps and are critical in CNNs for reducing dimensions.”

—