Question 1

Why is it important to iterate on data visualisations based on feedback?

Accepted Answer

Iteration allows for refinement and improvement of visualisations, ensuring they effectively communicate insights to the intended audience.

Question 2

Can clustering algorithms handle categorical data?

Accepted Answer

Some clustering algorithms, like k-modes for example, are designed to handle categorical data, but preprocessing may be required for others.

Question 3

How does leave-one-out cross-validation differ from k-fold cross-validation?

Accepted Answer

Leave-one-out cross-validation is a special case of k-fold cross-validation where each fold consists of a single data point, providing a more thorough assessment but at a higher computational cost.

Question 4

Can you use PCA for feature selection in machine learning?

Accepted Answer

PCA is primarily used for dimensionality reduction, but it indirectly contributes to feature selection by identifying principal components that capture the most variance in the data.

Question 5

What role does the learning rate play in training neural networks?

Accepted Answer

The learning rate determines the size of the steps taken during optimization. Finding an appropriate learning rate is crucial for achieving optimal convergence and avoiding overshooting or slow

data science techniques pdf

10+ Most Commonly Used Data Science Techniques in 2025