Machine Learning based Prediction of Internet Path Dynamics

We study the problem of predicting Internet path changes and path performance using traceroute and machine-learning techniques. Path changes are frequently linked to path inflation and performance degradation. Therefore, predicting their occurrence could improve the analysis of path dynamics using traceroute. By relying on neural networks and using empirical distribution based input features, we show that we are able to predict (i) the remaining life time of a path before it actually changes, and (ii) the number of path changes in a certain time slot with relatively high accuracy. We also show that it is possible to predict path performance in terms of latency, opening the door to novel, machine-learning-based approaches for RTT prediction.