proposal - Githubissues

UBC-MDS / DSCI_522_Group_18

Group project repo for DSCI 522

MIT License

0 stars 5 forks source link

Closed abaghela closed 3 years ago

abaghela commented 3 years ago

Some questionsn that come to mind when reading your proposal:

why chose 50K as the cut off?
why chose logistic regression? Can you compare this to other models? Which ones?
what measure of ML accuracy will you use? AUC, F1, or just plain accuracy?

abaghela commented 3 years ago

Also, the correlation network did not add much. Could probably have stayed in the EDA files.

AishwaryaGopal12 commented 3 years ago

Added reasoning for why choose 50K as the cut off in the introduction section of the report

AishwaryaGopal12 commented 3 years ago

We have addressed all the above points in the updated version