Unsupervised Learning and Principal Component Analysis Quiz

What is the primary goal of unsupervised learning?

• Prediction of a response variable

• Study relationships between variables

• Maximizing prediction accuracy

• Reducing dimensionality of data

Which of the following is a method used in unsupervised learning?

• Regression analysis

• Principal components analysis

• Time series analysis

• Classification techniques

What does PCA primarily aim to achieve?

• Create a response variable

• Reduce the dimensionality of the data

• Increase the number of variables

• Enhance prediction accuracy

In PCA, what are the new variables obtained called?

• Loadings

• Principal components

• Eigenvalues

• Response variables

What is a key characteristic of the principal components in PCA?

• They are correlated with each other

• They are linear combinations of original variables

• They are independent of the original variables

• They have no relation to the original variables

Why is unsupervised learning considered more challenging than supervised learning?

• It requires more data

• There is no response variable to validate results

• It is less computationally intensive

• It is easier to interpret results

What does the term 'standardization' refer to in the context of PCA?

• Creating new response variables

• Decreasing the number of observations

• Increasing the number of variables

• Transforming data to a uniform scale

What is the purpose of the scree plot in PCA?

• To check for multicollinearity

• To determine the number of principal components

• To visualize the original data

• To assess the accuracy of predictions

What does the proportion of variance explained (PVE) indicate?

• The share of variability explained by principal components

• The total number of variables

• The accuracy of predictions

• The correlation between variables

How are the coefficients (weights) for the first principal component determined?

• By averaging the original variables

• By maximizing the variance of the first principal component

• By minimizing the correlation with other components

• By standardizing the original variables