Recovery of Missing Information in Graph Sequences
Various procedures for the detection of anomalous events and network behavior were introduced in previous chapters of this book. In the current chapter we are going to address a different problem, viz., the recovery of missing information. Procedures for missing information recovery are useful in computer network monitoring in situations in which one or more network probes have failed. Here the presence, or absence, of certain nodes and edges is not known. In these instances, the network management system would be unable to compute an accurate measurement of network change. The techniques described in this chapter can be used to determine the likely status of this missing data and hence reduce false alarms of abnormal change.
KeywordsDecision Tree Feature Vector Leaf Node Linear Prediction Machine Learn Approach
Unable to display preview. Download preview PDF.