Some things I think are still missing from the report

[cwickham]

I know we are short in length but these are some things I think really need to be mentioned.

• F1 score interpretation: I don't have any intuition on what the F1 score tells us. Is it somehow comparable to any things like "positive predictive value", "specificity/sensitivity" etc.?

• OOB samples and error measures: I think you need to make it clearer in the text that OOB samples are used to pick the number of trees. This might mean you need to mention earlier in the random forest section what out of bag samples are.

• The discrepancy between error rates of OOB versus test set: come back to this in the Discussion and reiterate why this happens.

• Limitations / Future Work: I think in Discussion you should mention things you would have liked to have done but haven't yet, e.g. experiment with other tuning parameters.

[cwickham]

• You need to link to this GitHub repo somewhere in the text, and emphasize all the code to reproduce the analysis is here.
• Maybe this was somewhere and I missed it, but make sure to mention exactly which function and package you used to fit the random forests, along with a citation.