Abstract:
|
We study the problem of classification of networks with labeled nodes, motivated by applications in neuroimaging. Brain networks are constructed from imaging data to represent functional connectivity between regions of the brain, and previous work has shown the potential to diagnose disorders, giving rise to a network classification problem. Existing approaches to graph classification tend to either treat all edge weights as a long vector, ignoring the network structure, or focus on the graph topology while ignoring the edge weights. Our goal is to design a graph classification method that uses both the individual edge information and the network structure of the data in a computationally efficient way. We are also interested in obtaining a parsimonious and interpretable representation of differences in brain connectivity patterns between classes, which requires variable selection. We propose a graph classification method that uses edge weights as variables but incorporates the network nature of the data via penalties that promote sparsity in the number of nodes. The method shows good performance on data from two fMRI studies of schizophrenia.
|