After you've run a List of Items automatic categorization, you may want to tweak and further improve the results. This article describes how to review and use suggestions where the automatic categorization algorithms think that the data could potentially be tidied further:
The variant suggestions can be added automatically to further train and improve the automatic categorization:
Requirements
- A List of Items text analysis output, see How to Automatically Classify Lists of Items for instructions to create one.
Method
- Select your List of Items output on the Page.
- Click the arrow next to the Diagnostics section at the bottom of the output to expand it.
- You will see a Variant Suggestions section that contains suggested category variants. Copy the rows of variants by highlighting the suggestions with your mouse and press Ctrl+C to copy or right-click and select Copy.
- From the object inspector, go to Data > Required Categories > Edit required phrases and variants.
- A spreadsheet will pop up on your screen. Paste the variants using Ctrl+V into the spreadsheet, starting on row 2, like so:
- Click OK.
- The automatic categorization output will recalculate and include the variants that were pasted in. You may want to review Diagnostics > Variant Suggestions to see if further improvements can be made.
- To save the categories to use in other analyses, from the object inspector, go to Data > Save Variable(s) > Categories.
NOTE: The variables created from this using Save Variable(s) > Categories may become invalid and need to be deleted and recreated if the output has changed, either due to the input text variable being modified or the input settings being modified.
Next
How To Automatically Classify Unstructured Text Data Into an Entity List