Computer Algorithm Predicts Who Will Die This Season In Game Of Thrones

Christian Dallago/Technical University of Munich

Spoiler warning ahead, obviously, if you haven't seen seasons 1 to 5 of Game of Thrones...

One thing is certain about the new season of Game of Thrones: there’s going to be a few deaths. But while everybody is theorizing about what’s going to happen, a computer science class has taken their speculation to the next level.

Tutors and students at the Technical University of Munich in Germany have developed an algorithm that predicts whose neck is next up on the chopping block in season 6 of the popular HBO series. They also managed to work out which house is the most dangerous and discovered the public’s opinion on characters based on data from Twitter.

The program, which they aptly called “A Song of Ice and Data,” comes to its predictions after plugging in 24 different factors about each character, such as their age, their gender, their relations, and whether their parents are alive. And the most likely to die, according to the algorithm, is poor old king Tommen Baratheon, current occupier of the Iron Throne.

“This project has been a lot of fun for us,” said Dr. Guy Yachdav, who led the class and conceived the project, in a statement. “In its daily work, our research group focuses on answering complex biological questions using data mining and machine learning algorithms.

"For this project we used similar techniques. Only this time the subject matter was a popular TV show. The epic scale of the world created by George R. R. Martin provides an almost endless resource of raw multi-dimensional data. It provided the perfect setting for our class,” he added.

You can check out all of their predictions and insights right here. Perhaps most importantly, their algorithm also seems to be pretty clear on the fate of Jon Snow...

Comments

If you liked this story, you'll love these

This website uses cookies

This website uses cookies to improve user experience. By continuing to use our website you consent to all cookies in accordance with our cookie policy.