Artificial Intelligence Spam Prevention

Question

Is there any research / solution to prevent spam, bad language, etc. when entering a person using artificial intelligence methods, such as natural language processing, social intelligence, etc.

Answer 1

There are many such solutions, I will describe some of the most widely used ones.

• Bayesian spam filtering - a classic approach based on Bayesian probability . This is an old technique, so if you want to use it, consider different types of heuristics to improve results (see the Wikipedia article for more details).

• , LSA. LSA Random Indexing ~ 99,3% 30% . PLSA, LSA .

• . , , 2 , , , . , SVM.

Answer 2

- - . , . , Stackoverflow , CAPTCHA, , , .

Answer 3

Bayesian Network, .

.

Answer 4

ai-class.com. .

Answer 5

, AI Chatbot.