Improperly Handles hyphenated drug names (Fluorofentanyl vs Para-Fluorofentanyl

UK-IPOP / drug-extraction

A ToolBox for fuzzily extracting drugs mentions from text.

https://drug-extraction.vercel.app

MIT License

3 stars 0 forks source link

Improperly Handles hyphenated drug names (Fluorofentanyl vs Para-Fluorofentanyl #82

Closed Knoxort closed 5 months ago

Knoxort commented 6 months ago

We found that the term "Fluorofentanyl" had been overcounted in our results. Upon investigation, we found that the hyphen "para-fluorofentanyl" was imporperly handled, so that "fluorofentanyl" was matched with a simlalarity score of one. I've attached a screenshot of the output joined with the searched file as well as a CSV showing the whole file. Example of Hyphenation Error, Fluororfent vs P-Fluorofent.csv

nanthony007 commented 5 months ago

@trokon resolved w/ #83

New release followed.

Should be available (v1.3.0 in PyPI)