Learn R Programming

FedIRT (version 1.1.0)

Federated Item Response Theory Models

Description

Integrate Item Response Theory (IRT) and Federated Learning to estimate traditional IRT models, including the 2-Parameter Logistic (2PL) and the Graded Response Models, with enhanced privacy. It allows for the estimation in a distributed manner without compromising accuracy. A user-friendly 'shiny' application is included.

Copy Link

Version

Install

install.packages('FedIRT')

Monthly Downloads

186

Version

1.1.0

License

MIT + file LICENSE

Maintainer

Biying Zhou

Last Published

September 28th, 2024

Functions in FedIRT (1.1.0)

personscore

Federated IRT person score
runserver

Server for Federated IRT Model Estimation
runclient

Client for Federated IRT Model Estimation
fedirt_2PL

Federated 2PL model
example_data_graded

Graded Response Dataset for Federated Graded Model
SE

Federated IRT SE
fedirt

Federated IRT model
fedirt_gpcm

Federated Graded Response Model Estimation Function
fedirt_file

Federated IRT model
g_logL_gpcm

Gradient of Log-Likelihood for the federated graded Model
g_logL

Gradient of Log-Likelihood for the federated 2PL Model
logL

Log-Likelihood of the federated 2PL Model
logL_gpcm

Log-Likelihood of the federated graded Model
logL_entry

Aggregate Log-Likelihood Function for Federated Learning
mem

Memoization Function for Speed Optimization
example_data_2PL

Binary Response Dataset for Federated 2PL Model
example_data_2PL_1

Binary Response Dataset for Federated 2PL Model
example_data_graded_and_binary

Graded Response Dataset for Federated Graded Model
example_data_2PL_2

Binary Response Dataset for Federated 2PL Model
personfit

Federated IRT person fit
g_logL_entry

Aggregated Gradient of Log-Likelihood for Federated Learning