It was funny to see how people use the data of my last Lost post trying to solve whom Kate loves. I think I can help… Take a look at this new visualization focused in the Kate words. All she said from episode 1×01 to 6×11.

There are three different images. In the portrait the word size is proportional to word’s repetitions. The second one analyses the words around a determinate word, in this case the three possible winners, Jack, Sawyer and Aaron. The distance to the main word is the average distance. In the last one you can see the evolution of the 30 most used words by Kate along the show seasons.

Some considerations:
– I add “James” count to “Sawyer” in order to compare real totals.
– The size of the words around Jack, Sawyer and Aaron are proportionals to their respective numbers. That’s because I want to compare the relative importance to each one but don’t want to loose ‘resolution’ because of the ‘Jack’ big numbers.
– In the graphs, the value of each episode count are the average of episodies around it. That’s because in some episodies Kate don’t even appears.

You probably miss one word… but the only time she said ‘sex’ was when reading the Sawyer letter… in 01×08.

Enjoy it.

Download PDF JO-D-100413-KateEvolution-ING (709kB)

If you like this post you will probably like this others:
Jack vs Lock
Lost Characters
Get Lost

Uh, I forgot to mention… you can find the transcripts here.

Encouraged by this news, I made my own analysis of the last budget’s speech by Alistair Darling.

First I recreated the face of Mr. Darling with the data suggested. As usual the size of the words is proportional to the repetition.

Later, using my own text analysis over the last eleven budget’s speeches, I made different visualizations showing distance’s relationship and the evolution of the most used, or most significative, words.

The last one is focused in the word ‘tax’. In this case the distance to the words is the average distance in the speeches. The big one represent all years from 2005 to 2010. The other two are focused in their own year.

Due to the filesize I recomend you to download and navigate the PDF version.

Dwonload PDF JO-D-100411-Budget2010-ENG (621kB)