Dear colleagues,
I am trying to design a recurrent neural network to predict patients' length of stay as output. I have different types of data (numeric, categorical, free text, etc..) and I want to present the data as a time series( to capture the temporal dynamic of the system). It means I want the model to know when each piece of information is collected.
Here is my question:
1- What are the methods to feed the data into the model?
2- What are the advantages and disadvantages of each method?