Performing Principal Component Analysis of a set of tweets.
Hello! First and foremost, I apologize if this topic has been found somewhere. I have spent a considerable amount of time attempting to look for a method.
I have found 2 social science studies that utilized PCA of text data using Rapid Miner. They displayed in a table which words had the highest eigenvalue for a particular factors. I am interested in learning how to do this, but thus far I have been frustrated with a lack of process/steps. I also wonder if it is something so elementary that there are no methods that explain the process?
To be more specific, I am interested in analyzing an excel file containing 2000 tweets (for starters). Thank you in advance for your sincere assistance!