The results from my analysis show a low PPV (84.73%) compared to Sensitivity (88.8%). Not sure how to interpret these statistics, need some help

                 Manual Process
                  No           Yes
     New    No   3583          75
Automated   Yes  100           550

The interpretation depends on you data and your objective but generally, if your PPV is lower than your sensitivity, you have more false-positives than false-negatives in your data. Maybe you can provide more context, for example the specificity or false-positive-rate?

Sorry for not writing this as a comment but unfortunately I don't have enough reputation for that yet.

  • $\begingroup$ your suggestions were helpful :) , I have developed an algorithm that can automate the current existing manual review process which is very tedious, I have updated my question with a confusion matrix. $\endgroup$ – Ezra Polson Jun 24 '15 at 15:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.