COSNet (version 1.6.0)

runSubnet: Realizing the network dynamics.

Description

Function to simulate the dynamics of the network composed of unlabeled nodes.

Usage

runSubnet(W, labeling, alpha_value, c_value, cost)

Arguments

W

square symmetric named matrix, whose componemts are in the interval [0,1]. The i,j-th component is a similarity index between node i and node j. The components of the diagonal of W are zero. Rows and colummns should be named identically.

labeling

vector of node labels: 1 for positive examples, -1 for negative examples, 0 for unlabeled nodes

alpha_value

real value in [0, pi/2[, determining the neuron activation values: sin(alpha) and -cos(alpha).

c_value

real value used as activation threshold for each neuron

cost

real value corresponding to beta parameter in the equation eta = beta*|tan((alpha - pi/4)*2)|, where eta is the coefficient of the regularization term in the energy function (Frasca et al. 2013). If cost = 0 (default) the unregularized version is executed. The higher the value of cost, the more the influence of the regularization term. It is suggested using small values for this parameter (e.g. cost = 0.00001)

Value

list with three components:

state

Named vector of prediction (at equilibrium) for unlabeled nodes

scores

Named vector of scores (at equilibrium) for unlabeled nodes

iter

Number of iterations of the network until convergence

Details

Function to simulate the subnetwork composed of the unlabeled genes, in which each neuron has sin(alpha_value), -cos(alpha_value) as activation value, and in which each neuron has a threshold "c_value" minus the contribution from the labeled neighbors.

References

Frasca M., Bertoni A., Re M., Valentini G.: A neural network algorithm for semi-supervised node label learning from unbalanced data. Neural Networks, Volume 43, July, 2013 Pages 84-98.

Examples

library(bionetdata);
data(Yeast.STRING.data);
data(Yeast.STRING.FunCat);
n<-nrow(Yeast.STRING.data);
## removing dummy node 00
Yeast.STRING.FunCat <- Yeast.STRING.FunCat[,
                        -which(colnames(Yeast.STRING.FunCat) == "00")];
class <- 1;
labels <- as.vector(Yeast.STRING.data[, class]);
names(labels) <- rownames(Yeast.STRING.FunCat);
labels <- as.vector(Yeast.STRING.FunCat[, class]);
names(labels) <- rownames(Yeast.STRING.FunCat);
folds <- find.division.strat(labels, 1:n, 3);
labels[labels <= 0] <- -1;
test.set <- folds[[1]];
training.set <- setdiff(1:n, test.set);
labels[test.set] <- 0;
res <- runSubnet(Yeast.STRING.data, labels, alpha=1, c=0, cost=0.0001);
str(res);