That's it, now you can applaud.
The idea is simple: you need to find a right pattern that helps a driverless car to achieve its goal, and not to get into an accident. That is possible only knowing the explicit context and implicit subtext of the pattern. That is, by solving the problem posed by NIST TREC QA, creating a system capable of answer both Factoid and Definition (Other) questions.
I answered NIST TREC challenge by discovering and patenting AI-parsing, which allows structuring text and extract 100% of its information. (Only n-gram parsing was known before my discovery.) In addition, the same parsing helps to explain the pattern by its context and subtexts, annotating it with other texts.
That's it, now you can applaud.
Biting the hand that feeds IT
