Visualization of Multivariate Data DATA-SCIENCE-OPTIMIZE Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is a key advantage of a parallel coordinates plot compared to a scatterplot matrix?

   • AIt is easier to read when there are many data points.
   • BIt preserves the exact correlation coefficient between every pair of variables.
   • CIt can show relationships between more than two variables in a single plot.
   • DIt does not require scaling of variables.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A genomics researcher has a dataset with 2,000 gene expression measurements for 300 samples. The researcher wants to project the samples into a 2D scatter plot to see if there are natural groupings (e.g., by tissue type) while preserving the global structure of the data as much as possible. Which technique is most appropriate?

   • AA parallel coordinates plot with all 2,000 genes as axes.
   • BPrincipal Component Analysis (PCA) and plot the first two principal components.
   • Ct-SNE with a perplexity of 5.
   • DA scatterplot matrix of the first 5 genes.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A sports analytics team has performance data for 500 athletes across 8 numeric metrics (speed, agility, endurance, etc.). They want to identify clusters of athletes who share similar overall performance profiles and see how individual athletes compare across all metrics simultaneously. Which visualization should they use?

   • AA bar chart of the average of all 8 metrics per athlete.
   • BScatterplot matrix of all 8 metrics.
   • CA single heatmap of the 500 athletes by 8 metrics.
   • DParallel coordinates plot with each axis representing one metric.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A data scientist is analyzing a dataset with 100 features and 10,000 observations. The scientist needs to create a 2D visualization that preserves the global structure (distances between clusters) as faithfully as possible for a report. The dataset is known to have a roughly linear underlying structure. Which approach is most appropriate?

   • AApply t-SNE with a high perplexity.
   • BCreate a heatmap of the correlation matrix.
   • CApply PCA and plot the first two principal components.
   • DUse a parallel coordinates plot with all 100 features.
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   Which statement correctly describes how PCA (Principal Component Analysis) reduces dimensionality?

   • AIt creates new variables that are linear combinations of the original variables, ordered by the amount of variance they capture.
   • BIt selects a random subset of the original variables to keep.
   • CIt removes variables that have low correlation with the target variable.
   • DIt clusters similar observations together and uses cluster centroids as new variables.
   Reveal Solution Discussion Select an answer first