Contains: Writings about AP English readings and a little gadget of goldfish, that can be fed because, well, everybody needs swimming goldfish that can be fed with a click of a mouse on their blog. Does not contain: Really, anything other than those two things. I apologize for the lack of variety, but hey, interactive goldfish.
Monday, April 8, 2013
TED Talk
Erez Lieberman Aiden and Jean-Baptiste Michel talk about the creation of the Ngram in their "What we learned from 5 million books" TED Talk. Aiden and Michel began by asking if the well-known statement, "A picture is worth a thousand words," is true or not. They came to the conclusion four years later that a picture is definitely not worth a thousand words. It's worth 500 billion words. They did this by figuring out a way to analyze human culture through Google's scans of 5 million accredited, useful books. With these scanned books, they can search for a particular phrase or word and find out how many times it was used in a particular year and then graph all the years on the same graph, showing a trend of a particular word's use. Through these graphs, people can see what culture was focusing on in a particular time period and the influence of suppression and propaganda can be seen as well. Now, Aiden and Michel's Ngram technology is available to the public through Google. They quickly discovered that it was just as interesting to see what words people graph as it was to graph words themselves. In conclusion, Aiden and Michel set out to see if an old saying was right or wrong and instead discovered a way to graph cultural changes throughout the years.
"What we're left with is a collection of 5 million books, 500 billion words. A string of characters a thousand times longer than the human genome. A text which when written out would stretch from here to the moon and back 10 times over. A veritable shard of our cultural genome.
Considering the fact that over 129 million books have been published, Google has only scanned 15 million, and they're only considering 5 million of these 15 million. The true amount of characters of "the human cultural genome" would be multiple times more astounding than this. This is already an astounding amount of human culture, but it is absolutely no where near all of it.
The idea that no one cared about a year before it got there and that, progressively, people care less about a year sooner after it ends now than they used to is mind blowing. How would this graph look for years like 2000 where there was lots of anticipation for it? Or 2012?
Aiden and Michel's point about censorship with the artist Mark Chagall in English literature versus German literature was extremely interesting. The use of a famous person's name generally progresses in an upward direction, unless there's an outside force changing the data, such as censorship. Chagall, a Jewish artist, was censored in Nazi Germany, so his name wasn't mentioned at all at this point in German cultue. When Nazi Germany was defeated, his name became popular again. Outside influences on culture like censorship and propaganda show very clearly in the graphs of word usage as well.
"We found some pictures that are worth 500 billion words."
This concept is both extremely impressive and extremely confusing. How is a single picture worth 500 billion words? How does the original question of whether or not a picture is worth a thousand words connect to the rest of the talk?
It's interesting to see what everyday people type into the Ngram. What does this say about culture, that people find what other people are curious about interesting?
Aiden, Erez Lieberman and Jean-Baptiste Michel. "What we learned from 5 million books." TED. Boston, MA. July 2011. Guest Lecture.
Subscribe to:
Post Comments (Atom)

No comments:
Post a Comment