Training Positive/Negative Intents as separate intents

Hi,

I am trying to create 2 separate Intents based on positive and negative intents. What is the best way to train the Model, so that it differentiates negation in the data.

For example -

I like spanish food - positive intent

I do not like spanish food - negative intent

Is their a way to separate the intents based on negation in training data.

Thanks