fast_psa {svs}

Probabilistic Latent Semantic Analysis

Description

A fast procedure for computing probabilistic latent semantic analysis.

Usage

fast_psa(dat, k, symmetric = FALSE, transform = 1, tol = 1e-08)

fast_psi(dat, k, symmetric = FALSE, transform = 1, tol = 1e-08)

fast_plsa(dat, k, symmetric = FALSE, transform = 1, tol = 1e-08)

fast_plsi(dat, k, symmetric = FALSE, transform = 1, tol = 1e-08)

Arguments

dat

Input data: can be a table or a data frame (but the data frame must have only two columns).

k

Numeric specification of the number of latent classes to compute.

symmetric

Logical indicating whether to compute the symmetric or the asymmetric solution.

transform

Numeric specification of the "tempering" transformation as explained in Hofmann (1999: 51-52).

tol

Numeric specification of the convergence criterion.

Details

From version 1.1.0 of the svs package on, probabilistic latent semantic analysis is a special case of latent class analysis.

Value

A list with components:

prob0

The probabilities of the latent classes.

prob1

The probabilities for the first set of levels (viz. the row levels of a frequency table). The rows of prob1 sum to 1 if symmetric = FALSE, the columns sum to 1 if symmetric = TRUE.

prob2

The probabilities for the second set of levels (viz. the column levels of a frequency table). The columns of prob2 sum to 1.

References

Hofmann, Th. (1999). Probabilistic latent semantic indexing. SIGIR'99: Proceedings of the 22nd annual international SIGIR conference on research and development in information retrieval, 50–57.

Examples

SndT_Fra <- read.table(system.file("extdata", "SndT_Fra.txt", package = "svs"),
   header = TRUE, sep = "\t", quote = "\"", encoding = "UTF-8",
   stringsAsFactors = FALSE)
psa_SndT_Fra <- fast_psa(SndT_Fra, k = 7)
psa_SndT_Fra