Name
Abstract | ||
|---|---|---|
User activity in five mobile games is found to be accurately described by stochastic processes related to recurrent event models in survival analysis. We specify four simple parametric models and methods to fit them to data which specify this process within day accuracy in the individual user level. This model implies commonly used population level retention metrics: retention, rolling retention and lifetime retention. Furthermore, modelling aids in understanding the underlying phenomena generating these metrics, which is verified visually in five diverse mobile games. The model assists in obtaining analytical insight into frequency and longevity of product use and precipitates predictive modelling by forecasting their evolvement over time. |
Year | Venue | Field |
|---|---|---|
2016 | IEEE Conference on Computational Intelligence and Games | Population,Parametric model,Simulation,Computer science,Stochastic process,Artificial intelligence,Predictive modelling,Machine learning |
DocType | ISSN | Citations |
Conference | 2325-4270 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Markus Viljanen | 1 | 7 | 1.92 |
| Antti Airola | 2 | 763 | 39.79 |
| Tapio Pahikkala | 3 | 1005 | 70.68 |
| jukka heikkonen | 4 | 249 | 48.76 |