An OpenRefine reconciliation service for FAST.
FAST is available as Linked Data, which is an approach to publishing data which enhances the utility of information on the web by making references to persons, places, things, etc. more consistent and linkable across domains.
The service queries the FAST AutoSuggest API and provides normalized scores across queries for reconciling in Refine.
pip install -r requirements.txt
Run locally as:
$ python reconcile.py --debug
Michael Stephens wrote a demo reconcilliation service that this code is based on.
Instead of the score variable bringing back the relevance score of terms, I instead changed this to bring back the MARC field of the terms the API suggests.
It was necessary to change score because the extension currently only allows for a static number of variables, and score was the least useful for our purposes.
The MARC field tag helps us determine what type of terms we are dealing with, and even if we don't choose to use the FAST heading, we can organize them to reconcile to different sources like GeoNames, VIAF, etc.
I also edited the requirements.txt to include the python-Levenshtein pip package, as is recommended if you run the program without it.