Java – an algorithm for calculating the similarity between texts

I tried to score similar between posts on social networks, but I didn't find any good algorithms, ideas?

I've only tried Levenshtein, Jaro Winkler and others, but those people are more used to less emotional texts In the post, we can get a text "I really love dogs" and another saying "I really hate dogs". We need to classify this case as completely different

thank you

Solution

You may want to view opinion mining and sentimental analysis to understand the complexity of the task

Short answer: there is no "good algorithm", only mediocre algorithm This is a very difficult problem Good luck