Anomaly Detection of Dam Monitoring Data based on Improved Spectral Clustering

<p>In response to the abnormal data mining in dam safety monitoring, and based on traditional spectral clustering, this paper presents an anomaly detection method improved clustering. This applies a distance density adaptive similarity measure. The natural eigenvalue is introduced adaptively select neighbors of points, redefined be combined with k-nearest neighbor. Furthermore, shared neighbor adjust between monitoring samples according regional density. Moreover, considering distribution data, initialization clustering centers optimized both feature. can prevent algorithm from local optimum, better adapt non-convex dataset, reduce number iterations, enhance efficiencies detection. Taking slab as research object, experimental datasets are formed. Experiments these further verify that effectively discrete superior classical detection.</p> <p> </p>