Identification of Critical Components in the Complex Technical Infrastructure of the Large Hadron Collider Using Relief Feature Ranking and Support Vector Machines

This work proposes a data-driven methodology for identifying critical components in Complex Technical Infrastructures (CTIs), which the functional logic and/or system structure functions are not known due CTI’s complexity and evolving nature. The uses large amounts of CTI monitoring data acquired over long periods time under different operating conditions. identified as those condition signals permit optimal classification functioning or failed state. includes two stages: first stage, feature selection filter method based on Relief technique is used to rank according their importance with respect state; second stage identifies subset among highlighted by that most informative identification performed basis evaluating performance Cost-Sensitive Support Vector Machine (CS-SVM) classifier trained several subsets candidate signals. capabilities proposed assessed through its application benchmarks highly imbalanced datasets, showing performances competitive obtained other methods presented literature. finally applied Large Hadron Collider (LHC) European Organization Nuclear Research (CERN), experiments physics; criticality has been confirmed CERN experts.