Saturday, November 15, 2014

Baby Step


As I try to finish up my website, I am having trouble with a few things. Mostly time. It is amazing how much time it takes to do things on the computer! But on the positive side, I found a cool website for topic modeling called Regex. It doesn’t actually clean anything, but it can find and highlight anything you want to look for in your work. So if you wanted to double check for errors or just see what's in your documents or combine documents, this website would be helpful. It just won’t (as far as I know) take and replace anything. But OCR has been really difficult! And I haven’t even tried to clean PDFs! I’m just trying to get it done with a few word documents. But I think I might have finished my Mallet page. I still don’t really understand it, but my tutor helped me to go through some of the basics. I used the Command Line on the Terminal to look at three documents and see what it spit out. I still don’t really understand what all the numbers and how it’s set up, but it worked, I guess. I haven’t played with it enough to see how to actually use this for my research. The format doesn't make sense to me. Maybe I still need to put it into Excel or something. But I nonetheless have a screen shot (another new thing for me!) of the results. I also put it into Overview. That made even less sense, but maybe that’s because I only used three documents (and unrelated ones – just the three on top). Maybe if I play with it more and add more to it I will understand better. But the main thing is that I did it, and I even have pictures of my work. It’s a baby step, but a step! (Got to end these failure blogs on a positive note!)

No comments:

Post a Comment