Is there any categorization of datasets that contain simple(r) sentences?

[Colin Goldberg]

Hi,

For my purposes in developing/testing my approach, I would very much like to start with a dataset that contains simple/simpler sentences. The contexts (in the 2.0 devset) that I have inspected manually seem to comprise of (mostly) long or compound sentences. I would want to start with simpler sentences, and then move on to longer, compound sentences.

Is there any categorization that indicates the simplicity or otherwise of the sentences in any context in any dataset?

If not, I would like to recommend this for your consideration.

Thanks

Colin Goldberg