1- do data collection related to multi label examples.
each example can be classified into more than one class
2- create a classifier to read the data, then induce a classifier for each class.
3- use the micro averaging to calculate the error rate
we are interested in multi-label examples only
examples that can be classified into more than one class
You have to find the data set, build classifier and run the experiment
This is all the requirements, but we have some details:
like for example we want to know the parent-child relation
and the child-parent relation
also, it is very important to use the micro and macro averaging
to calculate the error
we need to run the experiment on more than one data set
it could be 4 or 5
1- do data collection related to multi label examples.
each example can be classified into more than one class
2- create a classifier to read the data, then induce a classifier for each class.
3- use the micro averaging to calculate the error rate
we are interested in multi-label examples only
examples that can be classified into more than one class
You have to find the data set, build classifier and run the experiment
This is all the requirements, but we have some details:
like for example we want to know the parent-child relation
and the child-parent relation
also, it is very important to use the micro and macro averaging
to calculate the error
we need to run the experiment on more than one data set
it could be 4 or 5