Principal Component Analysis in R

Course details

Dimensionality Reduction is a category of unsupervised machine learning techniques which  is used to reduce the number of features or variables of columns  in a dataset.

Lot of variables often enhances the noise signal in the data which is bad for modelling but Dimensionality Reduction techniques can help in this.

One of the Dimensionality Reduction Technique is Principal component Analysis which creates a new feature set which are uncorrelated or orthogonal .The newly created features are called Principal components.First principal component explains the most of the variance in the data and then the next principal component explains the remaining.

Principal Component analysis is helpful for any dataset which has many variables or variables which are anonymous.

Principal component analysis can help in explaining the structure of the dataset or creating the groups in the data or doing the predictive analytics .  

In this course we will discuss the following items:

• What is Principal component Analysis 
• When PCA is useful
• Data Visualization using  PCA in  2  & 3 Dimensions both
• What is principal component?
• What is the variance in the data in different dimensions?

• Properties of Principal Components
  1. Summarize PCA concepts
  2. Understand significance of first few PC's 
• Conduct PCA using R: 
• 1. Scaling/Normalization/Transformation
  2. Correlation Matrix
  3. Scree plot
•    Conduct k-means Clustering after Principal Component Analysis
• Visualizing the clusters in Reduced space