Visualization of Multivariate Data DATA-SCIENCE-OPTIMIZE Practice Questions (Page 6)

Questions 26–27

1. 26application · medium

   A data analyst has a dataset with 5 numeric variables and wants to explore the relationships between them. The analyst is considering a scatterplot matrix and a correlation heatmap. Which statement correctly describes a key trade-off between these two visualizations?

   • AA correlation heatmap shows the raw data, while a scatterplot matrix only shows summary statistics.
   • BA scatterplot matrix is better for more than 10 variables, while a correlation heatmap is better for fewer.
   • CA scatterplot matrix shows the raw data and non-linear relationships, while a correlation heatmap only summarizes linear relationships.
   • DBoth visualizations show the same information and are interchangeable.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the primary goal of dimensionality reduction techniques like PCA and t-SNE?

   • ATo increase the number of variables to improve model accuracy.
   • BTo convert categorical variables into numeric variables.
   • CTo project high-dimensional data into a lower-dimensional space while preserving important structure.
   • DTo remove all missing values from the dataset.
   Reveal Solution Discussion Select an answer first