For this experiment with the Topic Modeling Tool, I uploaded my corpus from Lab 3. Here is a screenshot of the topic index using the .html data from the tool from my time in class.
This list represents the most common topics found in the corpus I uploaded. For this iteration, I had the topic modeling tool show me 15 topics, as listed in the picture above. When I click on one of the topics, I am shown the top-ranked documents in the clicked topic. Here is a shot of the top ranked documents in the first topic.
Next I clicked on the first piece of work in this list, which is Once A Week by A.A. Milne. This more in-depth look shows us the percentage of words in this document assigned to each topic it falls under. Or, in other words, the weight of each topic present in the document. The percentages for each topic in Milne’s piece are shown in the screenshot below.
I can use this data from the tool to look at all of the other topics and all of the other pieces in my corpus. I decided to focus instead on the data and the data analysis of the Topic Modeling Tool instead of the actual using of the tool since everyone did it together and had no difficulty with it so all of the posts will be identical. By using my own data, I felt the post would be more interesting and would better show my understanding of the tool. I also included a screenshot of the command line we worked on in class, which I had trouble with initially only because my computer decided to crash as soon as I tried to open the program.
The command line shows another way to operate a computer besides using the user interface. This command line exercise taught me how to jump from folder to folder to desktop and also taught me how to find out exactly where I was in the computer and what existed in the location I was in.