MentDB Strong has a deep understanding core. It's a system that I invented that allows to create machines more than one-shot. Google Assistant, Siri, Alexa are one-shot machines. That is to say, they do not detect the ambiguities. They just look for a single answer (as close as possible to what the user wants), and stupidly return the answer. MentDB Strong does that too, but takes it deeper in the spirit. Indeed, the machine can at the slightest ambiguity, lock the mind in order to receive from the user a confirmation or a multiple choice. It is therefore possible to know the level of confidence in understanding if we know how the mind map works, and to create machines that reconsider what they have understood.
Different level of deep understanding
MentDB Strong can handle multiple levels of deep understanding. For the 1st level example: the fact that a strategy is exactly the same as the one just before. Made an addition with 12 and 99. 88 and 77, 10 and 55 ... Another level: what I say will be used for a task / thought on hold / locked. Give me the first word of a sentence. John is tall. the answer is John. I counted 22 possible locks so far in my version 4 of MentDB Strong. And there are others. In the years to come, thanks to these locks, I will be able to create machines that rarely go wrong, and when they do, they can go back. Like a human.