t-step Markov Random Walks

Share:

Description

t-step Markov Random Walks

Usage

1
2
sslMarkovRandomWalks(xl, yl, xu, t = 10, dist.type = "Euclidean", k = 10,
  gamma = 1, improvement = 1e-04)

Arguments

xl

a n * p matrix or data.frame of labeled data.

yl

a n * 1 binary labels(1 or -1).

xu

a m * p matrix or data.frame of unlabeled data.

t

step size.

dist.type

character string; this parameter controls the type of distance measurement.(see dist or pr_DB).

k

an integer parameter controls a k-nearest neighbor graph.

gamma

a numeric parameter in the affinity matrix.

improvement

numeric. Maximum allowed distance between computed parameters in two successive iterations at the steady state.

Details

sslMarkovRandomWalks transmits known labels to unlabeled data by t-step Markov random walks.Parameters are estimated by an EM algorithm.

Value

a m * 1 integer vector representing the predicted labels of unlabeled data.

Author(s)

Junxiang Wang

References

Szummer, M., & Jaakkola, T. (2001). Partially labeled classification with M random walks. Advances in Neural Information Processing Systems, 14.

See Also

pr_DB dist

Examples

1
2
3
4
5
data(iris)
xl<-iris[c(1:20,51:70),-5]
xu<-iris[c(21:50,71:100),-5]
yl<-rep(c(1,-1),each=20)
yu<-sslMarkovRandomWalks(xl,yl,xu)