Learn R Programming

COSNet (version 1.6.0)

Cost Sensitive Network for node label prediction on graphs with highly unbalanced labelings

Description

Package that implements the COSNet classification algorithm. The algorithm predicts node labels in partially labeled graphs where few positives are available for the class being predicted.

Copy Link

Version

Version

1.6.0

License

GPL (>= 2)

Issues

1

Pull Requests

0

Stars

1

Forks

2

Maintainer

Marco Frasca

Last Published

February 15th, 2017

Functions in COSNet (1.6.0)

generate_labels

Generate random labels
runSubnet

Realizing the network dynamics.
COSNet

Cost Sensitive Network for node label prediction on graphs with highly unbalanced labelings
find.division.not.strat

Random partitioning of input data
reg_data

Function to compute the regularized version of COSNet (Frasca et al. 2013)
optimizep

Optimizing algorithm parameters
generate_points

Compute the points to be separated in Step 2 of COSNet algorithm (Frasca et al. 2013)
COSNet-package

R package for binary classification and node ranking on partially labeled graphs with unbalanced labels.
cosnet.cross.validation

Cross validation procedure for the COSNet algorithm
find.division.strat

Compute a stratified random partition of the input data
optimize_pos_above

Alternative algortihm for optimizing parameters