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

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!


Lisa Han -Text Analysis Exercise

Page history last edited by Lisa Han 9 years, 7 months ago

I wanted to start out with something basic, so I went over to Google Ngram to check out how our usage of terms to describe Internet has changed over time. Nothing too surprising but it was cool to see it graphed out: “web” comes out on top but generally spikes around 2000. “Internet” lags behind by quite a bit, and it seems “cyberspace” has been fading out of use.

Next, I wanted to switch gears and compare some longer texts. I decided to go with the Battlestar Galactica pilot and try to compare it to a later season episode to see how the language changed. At first, I tried to do with this Divitext and treeview but I pretty come up with nothing useful. I switched over to Docuburst and entered “message” as the suggested root word. What I gleaned from this is that there is clearly a lot of crossover in language, but the pilot episode is a lot more dramatic—words like “problem” and “subterfuge” stand out as opposed to words like “truth” and “speculation” from season 4. Obviously I didn't clean the text well enough because there appears to be some spammy ad content here ("press execute etc).


I then took that pilot episode and mapped it out on textexture and got the following interactive visualization:

My final experiment was with sentiment analysis. I grabbed a tweet about Apple’s CEO Tim Cook and entered it into the demo to see whether it was negative or positive. I think the conclusion was that it was mostly neutral, so not too interesting there. The sentence: “Cook says if someone asks for FaceTime records, Apple can't supply it. And iMessage is now further encrypted too, he says.” –Casey Newton






Comments (0)

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