I have a set of tags (~10'000, will be extended over time) presented to a user. After he has selected 3 or more tags, I want to predict for each remaining tag what the chances are that the user will select this tag as well. I strictly need the prediction for all remaining tags. The prediction should be fast but accuracy isn't that important. The training data would be cases where users already have selected a subset of tags. What would be a good approach here?
welcome to ML and data science. This is a classic situation where a RNN would be usefull. You could either train one from Keras yourself which would help you learn and get a better model but might take more time. Or you could use a more pre-made soloution if this is your first project maybe something like TextGenRNN (will still recquire a bit of modification).