Detection of building structure damage with support vector machine

An important objective of health monitoring systems (HMS) for tall building is to diagnose the state of the building and to detect its possible damage. To solve these problems, data mining approaches are becoming meaningful along with the advance of Big Data techniques, among which support vector machine (SVM) is one of the most powerful classifiers because of its good accuracy. However, SVM is not suitable for large data sets or data stream classification which is the case of building structure data. In this paper we propose an online version of SVM for structural health monitoring. We construct a lab scale prototype, data collected from it were used to validate our approach. Experiment results show that the proposed SVM can detect the damage successfully, without a modeling process as traditionally people of the field do.