Next step, could be to know how to put names asociated to those alternatives. And classifying them by those names. That might help the search of such words when asked.
Not to be negative, but, why? The reason this is just sifting old nodes, as opposed to me/us building a hardcoded parser, is that it can be expanded with such things as neural nets to match different kinds fo patterns... without us having to step in and create useless things like names. It's the association of patterns that matters, and having some kind of name property would only add overhead. Unless I misunderstood what you meant by names and classifying by names?
That said, I do envision this as a kind of search tool. Besides returning the 'corrected' (we hope) code, it should give a list of answer-nodes which were relevant (in its estimation). | [reply] [Watch: Dir/Any] |
| [reply] [Watch: Dir/Any] |