Practice multiple choice questions on Principal Component Analysis (PCA) with answers. This is a fundamental technique in Machine Learning applications.
If you want to revise the concept, read this article 👉:
Let us start with the questions. Click on the right option and the answer will be explained.
Principal Component Analysis (PCA) is an example of?
a) Supervised Learning
b) Unsupervised Learning
c) Semi-Supervised Learning
Answer: b) Unsupervised Learning
Principal Component Analysis (PCA) is an example of Unsupervised Learning. Moreover, PCA is a dimension reduction technique hence, it is a type of Association in terms of Unsupervised Learning. It can be viewed as a clustering technique as well as it groups common features in an image as separate dimensions.
What is the importance of using PCA before the clustering? Choose the most complete answer.
Find which dimension of data maximize the features variance
Find good features to improve your clustering score
Avoid bad features
Find the explained variance
In order to improve your clustering efficiency, you need to find which dimension of data maximize the features variance, also with the result you can find the explained variance for each dimension. As more variance you have, less data you loss.
Following the steps to run a PCA's algorithm, why is so important standardize your data?
Find the features which can best predicts Y
Standardize data allows other people understand better your work
Use the best practices of data wrangling
Make the training time more fast
When you standardize your data, you will find the features with higher variance, and in this case, features that will better predicts our target (y).
What PCA does afterfall?
Reduce dimensionality of the data and create new features from features set given
Predicts your target with high efficiency
Create clusters in order to let you know what are the class
Give you the highest number of features possible, to maximize the efficience of your Machine Learning algorithm
With a reduced dimensionality it is easier to visualize your data using a clustering method, will help you to reduce noise from your data and the training phase will be faster.
Why you have to drop unimportant features?
Using the most important features will give you better efficiency predicting your target
Standardize the data
Find the correct clusters
To trains the model faster
Using only the most important features will give you a environment with less data loss, will let you know exactly which features is most important and will reduce dimensionality. All of this are the real purpose of PCA.
When uses PCA?
You want to find latent features and reduce dimensionality
Everytime before uses a Machine Learning algorithm
When I have a overfit case
When my data is small and with a few features
PCA helps your to find latent features among all your data, can reduce your dimensionality for 1/10, making easier to visualize data and faster training because uses less hardware to run.
With these questions on Principal Component Analysis (PCA) at OpenGenus, you must have a good idea of Principal Component Analysis (PCA). Enjoy.