Indeed it's one of the most difficult challenges I am facing, the solution would be to give the power to the users so that they can discard such parsing. But first I have to ensure that there are people who might be interested in using the solution, it's the phase I am in right now (trying to validate the concept), thanks for the feedback :)
It is definitely something I will explore if I continue the project, combining time and location.
I agree it is difficult to find locations for text but nowadays the NLP algorithms are more powerful than ever before so it is feasible :)
That was the original idea, it seemed too complicated to do at the beginning but now after working some months on the idea, I think it's feasible, I will add this feature in my future surveys to see who might be interested, thanks for the feedback, much appreciated :)
Thanks a lot for all the feedbacks, I agree with you, the more I work on the project the more I realize how difficult it is to make everything fully automated without impacting the quality of the geolocation (NLP is not 100% reliable) and the quality of the sources, I am still working on that.
I see your point, it's something listed in my list of bugs to fix, for now i am still in an idea validation phase where I want to validate the concept before committing more time to it, but I agree, this should definitely be fixed first :)
Thanks for the feedback, I will add the cities/ general areas, it's necessary indeed.
And sorry for the inconvenience, there was a problem with the database, everything is back to normal now.