5.0 / 5
We will cover the following tasks in 36 minutes:
In this section we will discuss course objectives, the Rhyme Interface and a short bio about me. Come in and see what this one is all about.
Story and Scraping Time
Really? Red Light Therapy? Ok. Well in this section we’re going to discuss how this analysis came about and we will scrape the data (326 Reviews) using Rvest and the tools we used in the previous Data Harvesting using Rvest course.
Cleaning the Data
Smoke is to fire what Dirty is to Data. Clean data is a dream. More often than not we have to clean it. This is what makes us invaluable to our teams, to our analysis, etc. In this lesson I will show you some down and dirty base R ways of cleaning data up.
More Cleaning with a For Loop
When you can’t think of a function off hand or are unaware of one use a For Loop. We will use a For Loop in this section to strip out empty columns. Sure there are other ways, but this is good practice. If you’re not familiar with For Loops it’s ok. See what you can learn and google what you can’t.
It’s time to take the Reviews and break them into single words. We do this so we can strip out extraneous words and those that don’t add have much meaning. We will be using the Reshape2 package to get this done. Come in and join me on this one.
In this lesson we will go over the Sentiment Libraries. Cool stuff. We will then join our words with the bing sentiment library using inner_join from the dplyr package. Once you see how this is done it will open all sorts of doors to your text analysis.
In this lesson we will build our comparison cloud using the wordcloud package. But first we need to separate the negative words from the positive ones so we can build the comparison part of the cloud.
Our analysis was simple but it provides a framework for future tutorials and your future projects. Bring a pen and paper to write down a very important book by Julia Silge, which is online and free. And of course you are learning a ton! And for that I want to congratulate you!