Version 3 (modified by Neil, 16 months ago) (diff)

add talk

Meeting 2019/08/17


Saturday, 17th August 2019, starting at around 14:00.


Superbalist have kindly offered to host us.


Gordon Inggs will give a talk titled "Making sense of a Large Organisation using spaCy and scikit-learn"


Recently, I was asked to help understand which jobs at the City of Cape Town are particularly "data"-intensive. After several fruitless keyword searches, we turned to Data Science to help us with our search. Using the excellent spaCy library, within minutes we were able to embed the City jobs into a high dimensional space. This allowed us to not only perform relevance searches which defeated those pesky synonyms, but also set things up for some decent data mining.

Come to the talk to find out all of the details, and maybe even help me help the City understand itself better.

Next Meeting

To be confirmed, but likely late September

python3 -c 'import datetime as D; print(, month=8, day=17) + D.timedelta(weeks=6))'