On November 26th, we had our 16th Amsterdam Girl Geek Dinner.
Karina van Dalen-Oskam showed us the link between literary science and computation. This territory is for a large part unmined. Quality reading is still the leading method for analysing texts but things are changing. Interestingly, a lot of tools are available for analysing newspaper texts, but they are not always working for literary works. A lot of manual labor is still needed to analyse literary works with these tools.
Karina talked about her project of analysing the use of names in novels. Children’s books use more personal names than any other genre.
The styles of different authors are distinguishable. Karina told us about the medieval book “Roman van Walewein” that had two authors and two scribers. At one point, one writer took over from the other. You can see this happen in a graph of word usage per line. Interestingly, the moment where the second scriber took over from the first can also be seen clearly.
Literary quality may be indicated by the number of different words used, and by the number of syllables. There is debate of how literary quality must be defined; is it in the eye of the beholder or is it a characteristic of the work itself?
Just 2 months ago, “Het Nationale Lezersonderzoek” was held, a big survey in which people could rank books and judge whether they were literary works or not. Unsurprisingly, 50 Shades Of Grey was at the bottom of the list ;)
Karina’s slides are here. On the last slide, Karina added some interesting links.
Post written by Marjon Wiendels.
Photos were taken by Coline Pannier.
Sponsor of AMS GGD #17
Sponsor of AMS GGD #16
Error: Twitter did not respond. Please wait a few minutes and refresh this page.