Recall & Review
beginner
What is the main goal of dimensionality reduction in data visualization?
The main goal is to reduce the number of features (dimensions) in data while keeping important information, so we can visualize complex data in 2D or 3D plots easily.
Click to reveal answer
beginner
Name two popular techniques used for dimensionality reduction visualization.
Two popular techniques are Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE). PCA is linear and fast, while t-SNE captures complex patterns but is slower.
Click to reveal answer
intermediate
How does PCA reduce dimensions?
PCA finds new directions (called principal components) that capture the most variance (spread) in the data, then projects data onto these directions to reduce dimensions.
Click to reveal answer
intermediate
What is a key difference between PCA and t-SNE for visualization?
PCA is a linear method that preserves global structure, while t-SNE is nonlinear and focuses on preserving local neighborhoods, making it better for visualizing clusters.
Click to reveal answer
intermediate
Why is it important to scale or normalize data before applying PCA?
Because PCA is sensitive to the scale of features, scaling ensures all features contribute equally, preventing features with large values from dominating the principal components.
Click to reveal answer
Which dimensionality reduction technique is best known for preserving local data structure in visualization?
What does PCA use to reduce dimensions?
Before applying PCA, why should data be scaled?
Which visualization dimension is most common after dimensionality reduction?
Which method is faster but less effective at capturing complex patterns?
Explain how dimensionality reduction helps in visualizing high-dimensional data.
Compare PCA and t-SNE in terms of their approach and visualization use.