How to create a multi label classification network in Keras if I have the training data with various accuracy?

I'm trying to create a neural network that finds the most effective treatment for each patient.

I have a medical database for training. The inputs are histological and pathological data (mostly 0/1 data of having some diseases, conditions) and the outputs are the treatments. And there is an accuracy score, which is the patient's reaction to the treatment (this shows how good was the treatment for the symptoms, so how accurate was the input-output pair).

How can I create a multi label classification network that finds the best treatment if the training data pairs are not always totally correct, rather having these variable accuracy scores?

Topic keras tensorflow neural-network python machine-learning

Category Data Science

1
answered at

One approach is to have a dummy class that represents no treatment and use the accuracy score via a threshold (lower than threshold) correspond to no treatment at all.

Threshold as used above becomes a new hyper-parameter and you have new input-output pairs that are now exact (depending on threshold).

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.