If we (okay, I... and the future version of my self) are going to categorize messages, we are going to resort to machine learning algorithms at some point. So we might as well do so from the beginning.
Potential categories:
Ideas
Quotes from articles I read on my phone
TODOs and notes to self
Things to check out later (movies, series, books, etc.)
Is there an exact match in google search (since the text can be long, choose a couple of shorter sub-texts and do exact search using Google search API)? One can argue that no ML is needed to detect quotes and indeed they would be right, but I think having it as one of the features fits nicely into a more generalized model.
Verbs in imperative mood and their positions. Is there a library which will make such checks easier?
If we (okay, I... and the future version of my self) are going to categorize messages, we are going to resort to machine learning algorithms at some point. So we might as well do so from the beginning.
Potential categories:
Possible features for ML algos: