# How to create a classification model for multi output dataset?

I have a dataset where there are two target variables target-1 and target-2.

Both target variables are ordinal and thus I want to develop a multi-class classification model over these two target variables. However, I am unsure as to where to start in the model creation process. Therefore, if someone could give some pointers to this, that would be greatly appreciated.

• Thank you for your question. Just to clarify you have a dataset (could you edit your post and include examples of your input?) and you want to predict a class for each of the target variables? Or are the target variables the classes themselves? – shepan6 Aug 4 at 9:04
• Hi, I have added a few data points of the dataset. There "breed_category" and "pet_category" are the target variables. I have experience working on single dependent variable but have no experience working on a multi-output variable dataset. So my question here is what process should be followed to create a classification model. The two target variables are multi-class variables so I would prefer classification model creation. – Kartik Rayaprolu Aug 4 at 9:11

So the question asks how to create a model such that you take in input X and translate this into two simultaneous predictions, one which predicts the breed_category and pet_category.
Depending on the number of the classes within the two targets (breed_category and pet_category), you could simply concatenate the outputs as one-hot encoded vectors (https://machinelearningmastery.com/why-one-hot-encode-data-in-machine-learning/).
So, you would encode the categories for breed_category and pet_category into one-hot encoded vectors and then concatenate them together.
If the number of classes from both targets is large, it might be better to have separate feedforward models, one for breed_category and pet_category.