Ana Macanovic amacanovic
amacanovic
/ ldatuning_keyATM.R
Last active
November 10, 2022 07:27
A code that allows you to use the measures from the ldatuning R package (https://github.com/nikita-moor/ldatuning) for use with keyATM models (https://keyatm.github.io/keyATM/). These measures allow you to choose the optimal number of topics to cluster textual data into with the keyATM models by Eshima, Imai, and Sasaki (http://arxiv.org/abs/200…
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Adapting the ldatuning functions to keyATM model structure in R | |
| # Code modified by Ana Macanovic (see amacanovic.github.io for contact details) | |
| # Please let me know if you find errors or have suggestions regarding these functions. | |
| # The function metrics_ldatuning below adapts the metrics from the ldatuning | |
| # package (https://github.com/nikita-moor/ldatuning) so they can be used with the output | |
| # of keyATM models (https://keyatm.github.io/keyATM/) . | |
| # See this discussion for more details: https://github.com/keyATM/keyATM/issues/169 | |
| # Note that the authors of keyATM models suggest that one does not need to adjust the number of topics |