Hello and welcome to Eurovision By The Numbers. A blog where I, a completely unqualified Australian, try to predict the outcome of the 2015 Eurovision Song Contest via statistics and machine learning, and hopefully learn them on the way.

So why am I doing this? Well firstly I love Eurovision. I am also a complete numbers geek, and have been meaning to try to learn about machine learning, and statistics for a while. When it was announced that Australia would be participating in Eurovision 2015, it seemed like the perfect opportunity to funnel my obsession into something vaguely practical. Writing a blog seemed like one of the best ways to keep myself accountable, and make sure I finish the project.

This is purely a passion project. I really should stress that I really do not know what I'm doing. I'm okay at math, but I basically just started learning R (or any sort of code!), learning about machine learning about a month ago. However since then I have been diving into the literature, both learning how to actually create a model, and learning all I can about Eurovision.

I will try to update the model once every week, with new data and improvements that I've made. The model is still very much in development, and is definitely likely to change as I find more data and work on it. I'll also try to post interesting findings from the data, or interesting relationships I've noticed. While I am obviously interested in predicting the winner, I am just as much focused on predicting the voting patterns for each country and I hope to have a model which will predict the full results.

The very lofty and unachievable goal I have set for myself is to try and create a model that is as accurate as the betting market. Lets see how it goes!