Raw to LDA make several assumptions on pre-processing rules. See the topic_flow_creation vignette for more details.

rawToLDA(folder, k, months)

Arguments

folder

Folder variable containing all raw data. See loadFiles.

k

The number of topics k for the model.

months

A character vector containing the months of interest (currently one uses the first month)

Value

A list structure containing 'tokens', 'dfm', 'LDA', used by various functions in the package.