Demystifying Principal Component Analysis

Data visualization has always been an essential part of any machine learning operation. It helps to get a very clear intuition about the distribution of data, which in turn helps us to decide which model is best for the problem, we are dealing with. Currently, with the advancement of machine learning, we more often need to deal with large datasets. The datasets are having a large number of features, and can only be visualized using a large feature space. Now, we can only visualize 2-dimensional planes but, visualization of data is also seems pretty necessary, as we saw in our discussion above. This is where Principal Component Analysis comes in.