• If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Stop wasting time looking for files and revisions! Dokkio, a new product from the PBworks team, integrates and organizes your Drive, Dropbox, Box, Slack and Gmail files. Sign up for free.


Ashley Jeun - Topic Modeling

Page history last edited by Ashley Jeun 4 years, 11 months ago

I downloaded the Topic Modeling Tool, and was able to see certain topic lists generated although I was confused on some parts of the program.


(1) I was unsure about the output/input parts of the tool, so I experimented by using my English 197 course folder as the input. I then used the text files ("Persuasion," "Attila," "Eric Brighteyes," "The Little Duke," "Man of the Forest," and "War of the Worlds"), which are saved in a separate folder, as the output option.


(2) Next I clicked on Advanced options to download a stop words list, while keeping the number of topics at 20. I left the no. of iterations, no. of topic words printed, and topic proportion threshold as they were.



(3) The lists I attained seemed very clerical and computed, and I realized I may have made the mistake of including certain files in my input that still had html codes. Some topics contain letters that don't even create a word. 


(4) In generating my topic lists, I haven't succeeded in creating accurate ones that I could understand. I also would like to clarify the purpose of the "tokens" which are recorded before each group of lists.

Comments (0)

You don't have permission to comment on this page.