Learn R Programming

topicmodels (version 0.2-17)

perplexity: Methods for Function perplexity

Description

Determine the perplexity of a fitted model.

Usage

perplexity(object, newdata, ...)

# S4 method for VEM,simple_triplet_matrix perplexity(object, newdata, control, ...)

# S4 method for Gibbs,simple_triplet_matrix perplexity(object, newdata, control, use_theta = TRUE, estimate_theta = TRUE, ...)

# S4 method for Gibbs_list,simple_triplet_matrix perplexity(object, newdata, control, use_theta = TRUE, estimate_theta = TRUE, ...)

Value

A numeric value.

Arguments

object

Object of class "TopicModel" or "Gibbs_list".

newdata

If missing, the perplexity for the data to which the model was fitted is determined. For objects fitted using Gibbs sampling newdata needs to be specified.

control

If missing, the control of the fitted model is used with suitable changes of the relevant parameters (see Details).

use_theta

Object of class "logical". If TRUE the estimated topic distributions for the documents are used. Otherwise equal weights are assigned to the topics for each document.

estimate_theta

Object of class "logical". If FALSE the data provided is assumed to be the same as the data used for fitting the model. The topic distributions therefore do not need to be estimated and the data in newdata is used for weighting the term-document occurrences.

...

Further arguments passed to the different methods.

Author

Bettina Gruen

Details

The specified control is modified to ensure that (1) estimate.beta=FALSE and (2) nstart=1.

For "Gibbs_list" objects the control is further modified to have (1) iter=thin and (2) best=TRUE and the model is fitted to the new data with this control for each available iteration. The perplexity is then determined by averaging over the same number of iterations.

If a list is supplied as object, it is assumed that it consists of several models which were fitted using different starting configurations.

References

Blei D.M., Ng A.Y., Jordan M.I. (2003). Latent Dirichlet Allocation. Journal of Machine Learning Research, 3, 993--1022.

Griffiths T.L., Steyvers, M. (2004). Finding Scientific Topics. Proceedings of the National Academy of Sciences of the United States of America, 101, Suppl. 1, 5228--5235.

Newman D., Asuncion A., Smyth P., Welling M. (2009). Distributed Algorithms for Topic Models. Journal of Machine Learning Research, 10, 1801--1828.