"Is it possible to detect offensive language on twitter, then block the user responsible?"

Hi,

Detecting the offensive language should be doable and you have at least two approaches:

• Either you have a collection of previously collected and labeled tweets (offensive/not offensive) and train a classifier 

or

• You query an API for you to do that (one of the first hits on google is that one: https://www.paralleldots.com/abusive-content)

As an idea how you can start building something like that, here is a recent research paper on exactly that topic:
https://arxiv.org/abs/1801.04433

Best,
David

@David_A Thanks, will have a go.