If you’ve ever played video games on your smartphone, you know how easy it is to get bored and switch to a new one. This is bad news for the video game industry, so researchers developed an algorithm that can predict when a user will get bored with a mobile game and leave it.
The new algorithm works using mathematical principles and a so-called “ensemble” method, which is based on many learning algorithms instead of just one. By examining and comparing many correlations and alternative models, this method greatly improves the prediction accuracy. In fact, according to África Periáñez who is Head of Game Data Science at the video game company called Silicon Studio, every time the researchers run this model, they’re actually using 1,000 different submodels.
That’s not all. Within each sub-model, the researchers used a survival analysis algorithm, which is something mostly used in medical research and biology, as it can predict when a patient will experience something important, or how certain cells are going to behave in a patient’s body.
Using a combination of all of these things, the model, when applied to video games, can predict what day and at what stage of the game a user will stop playing. This is actually great news for gamers, as the industry will surely take advantage of this algorithm to develop more entertaining and stimulating games that will occupy users’ attention for longer periods of time.
FECYT – Spanish Foundation for Science and Technology via ScienceDaily (https://www.sciencedaily.com/releases/2017/03/170324083031.htm)