I am currently working on a neural machine translation project and want to gather a corpus (or dataset) of internet texts that are written in standard and plain language. In theory, it certainly makes sense to try to collect all texts and compile them in a research corpus. In practice, I would like to proceed as systematically as possible in order to find at least the relevant or most suitable texts.
It is possible to find simple language texts relatively quickly, which can also be easily aligned. So the problem for me at the moment is not finding data, but rather finding a systematic way that allows me to prioritize the scraping and merging into a dataset. Ideally, the system is backed up by scientific literature.
At the moment I'm a bit stuck and would be happy to hear from you about known approaches, queries for literature research, known similar projects, specific literature or other such things :)