analysis/README.md

In nateaff/legolda: Color Topic Models of the Lego Dat Set

Analysis files

These files contain the analysis code and write-up. All R code except for a few library calls and initial data loading calls are in the files in the code directory.

Files

1. tf-idf Initial loading and munging of data. Computation and plots of TF-IDF and sample sets.
2. train-model.Rmd Contains all model validation code and plots. Cross-validation of perplexity scores, scoring of models with ldatuning functions, coherence scores and plots of set distributions over topic probabilities.
3. compare-model-distributions.Rmd Plots of color distributions over topics, plots of example sets and tables of example sets.
4. final-model.Rmd Several plots of the final model.
5. background Some math used in model evaluation.

nateaff/legolda documentation built on May 18, 2019, 10:15 a.m.