Class imbalance and classification metrics with aircraft wildlife strikes

In this screencast, I use the data set from the recent episode of SLICED I participated in, predicting damage from aircraft strikes with wildlife, to explore how handling class imbalance in modeling affects different classification metrics. Check out the code on my blog: juliasilge.com/blog/sliced-ai...

[@joel09878]

I've been learning tidymodels recently and it's remarkable how these videos resonate once you've done a little practice. Thanks again for the video and all your incredible work Julia

[@jonathanjayes]

Fantastic to see you on sliced and your update here Julia, thank you so much!

[@mohsinramay]

Thanks for spreading this useful information julia and keep us updating.

[@talitabac]

Fantastic video, thanks Julia!

[@darmaw22]

Once again, great video! Many thanks!

[@iqu3261]

Excellent and easy to understand method thank you Julia

[@AndyBarkerBee]

I have no idea why this video auto-played played straight after an Atlassian How To: Add Org Admin video... but hot damn, I am learning! WOW! Very cool video

[@No5Workshop]

This is interesting. This highlights the importance of setting an accurate goal, i.e. do we evaluate a model based on a single metric or a suite of metrics.

[@My-NaMeS_jEfF]

Julia, do you think the long plot would have been more useful if 'damaged' were a factor? Doesn't seem like our response should be character type.
Thank you!

[@benxneo]

Hey Julia! Thanks for your video, they are super helpful. I was wondering what you think about the language Julia, which is incidentally your name, have you dabbled with it?

[@JuliaSilge]

I haven't used it seriously enough to have a personal opinion, actually!