Algorithm predicts which 'Game of Thrones' characters will die

By Daniel Uria  |  April 22, 2016 at 2:15 PM
share with facebook
share with twitter
| License Photo

MUNICH, Germany, April 22 (UPI) -- A group of German students developed an algorithm that calculates the likelihood that various characters from HBO's hit series Game of Thrones will die.

The application, known as "A Song of Ice and Data" in reference to the collection of author George R.R. Martin's novels "A Song of Ice and Fire," searches the internet for information on each of the thousands of characters in the series and then calculates the likelihood of their death.

"In this project, we wanted to find features that are common to all already dead characters and use these features to predict the percentage likelihood of death (PLOD) for yet alive characters, aka to answer the question - who is likely to die next?" The project's website states.

The application was created by a group of students at the Technical University of Munich as a project for their computer science class and also includes an interactive map that tracks the movement of various Game of Thrones characters.

"In its daily work, our research group focuses on answering complex biological questions using data mining and machine learning algorithms," Dr. Guy Yachdav, who conceived the project said. "For this project we used similar techniques. Only this time the subject matter was a popular TV show. The epic scale of the worlds created by George R. R. Martin provides an almost endless resource of raw multi-dimensional data. It provided the perfect setting for our class."

The site lists Tommen Baratheon as the living character most likely to be killed with a likelihood of 97 percent, while Sansa Stark is listed as the top survivor with a death chance of only three percent.

It also keeps track of fans reactions to various characters on Twitter, with Jon Snow receiving the most positive mentions and Grey Worm receiving the most negative reactions.

Related UPI Stories
Trending Stories