Run and Optimize Cross-Situational Word Learning Models

A collection of cognitive models of cross-situational word learning, together with methods and datasets for fitting and cross-validating them against experimental and corpus data, and for plotting the results.

XSLmodels

R package for running and fitting cross-situational word learning models to data.

Installation

To install the latest development version:

# install.packages("devtools")
devtools::install_github("kachergis/XSLmodels")

Usage

See what models are available:

models <- show_models()

See what datasets are available:

datasets <- show_datasets()

Get model fits across all included datasets, for a given model:

group_fit_uncfam <- get_group_model_fit("uncfam")

Get cross-validated (5-fold) model fit:

cross_val_fit_uncfam <- get_crossvalidated_model_fit("uncfam")

Datasets held out from those fits -- for testing how well a fitted model generalizes -- are listed, with how to score each, in xsl_holdout_datasets.

For more details, see this vignette.

Models can also be run and scored against gold-standard lexicons from naturalistic child-directed speech corpora (rather than fit to lab experiment accuracy) -- see the corpora vignette.