Count Text Pairs
How to Access This Feature
Count pairs of tokens that appear simultaneously within a document.
From + (plus) Button
You can access it from 'Add' (Plus) button. "Text Mining..." -> "Count Text Pairs".
How to Use?
- Select a column that has tokenized text - Set a column that has tokens. This is "token" column if it's tokenized by do_tokenize function.
- Select a column as document id - A column considered as document id. If you run do_tokenize beforehand, this can be document_id.
- Keep Only Unique Pairs (Optional) - The default is TRUE. If FALSE, duplicated pairs appear in reverse order.
- Keep Diagonal Pairs (Optional) - The default is FALSE. If TRUE, count of the value itself appears.
- Sort the Result (Optional) - The default is FALSE. If TRUE, the output is in decreasing order of frequency.