All metrics to measure predictability are simply depend(Negatively) on False Negatives(FN) as well as False Positives(FP). Is there any way of knowing who(Instances/Tuples) went to FN basket as well as FP.Is there any backtracking of Modeling.
Md Hasan Al Banna
You can predict each sample and map it so that you know what samples were FN and what are the other samples.
0 votes
0 thanks
Muhammad Ali
Have a look at https://towardsdatascience.com/a-non-confusing-guide-to-confusion-matrix-7071d2c2204f
0 votes
0 thanks
Leonid Kupershtein
you must manually analyze results of your classification model and estimate which example take false value
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling
More Ramesh Eluri's questions See All
Similar questions and discussions