What technologies/techniques should be used for the service of automatic selection of interesting content?

Question

Python

I have a small script that selects interesting information from some VK publics. Now it works quite primitively: I set the keywords myself, and the script goes through the posts and searches (using regular expressions or fuzzy matching) for the necessary posts. But I understand that this method is very far from ideal, ineffective and all that. Therefore, I ask for help from Toaster's guru: what can be applied so that the system analyzes the posts it likes and learns itself? I suspect (but not sure) that machine learning should be used here. But there is absolutely no knowledge in this topic, so I will also be especially happy if you tell me the sources of information on the desired topic (preferably written ones, I don’t perceive videos well) in Russian or English.

Answer 1

https://github.com/checkcheckzz/python-github-proj...
the topic itself is immense, start, for example, with www.clips.ua.ac.be/pages/pattern of
course, the classic book
shop.oreilly.com /product/9780596516499.do
and a bunch of books on NLTK to a bunch of
well, and if you dig into the theory (there is a huge selection of material on everything), of course
www.dialektika.com/books/978-5-8459-1623-5. html
aka www-nlp.stanford.edu/IR-book
but I would make it easier: put everything in an index, for example, Sphinxsearch or ElasticSearch (gluttonous) and get used to returning results by relevance