Classification in Sparse, High Dimensional Environments Applied to Distributed Systems Failure Prediction

摘要: Network failures are still one of the main causes distributed systems’ lack reliability. To overcome this problem we present an improvement over a failure prediction system, based on Elastic Net Logistic Regression and application rare events techniques, able to work with sparse, high dimensional datasets. Specifically, prove its stability, fine tune hyperparameter improve industrial utility by showing that, slight change in dataset creation, it can also predict location failure, key asset when trying take proactive approach management.