Not to worry, Andy. I, like you, am just trying to be helpful.

I enjoyed your post in the “new language” poll, and I find your ideas intriguing, to say the least. I’ll reply properly on that thread, though, later this morning, when I’ve had the chance to assimilate the data (I just woke up, and I don’t think well, right off the bat).

AndyHo - Apr 24, 2011:

If you want to give a bot a real intelligence you must study Natural Language Processing, then master chunking, POS tagging, linguistic parsing, then go to the modelling of relations and information extraction, then decide a known model or create a new and possibly better one, and then incorporate to ot a model of knowledge, and fill some of it into the system (like the objects in the SHLDU room) Add a modal behavior and then the bot may answer.. stop, not answer.. simply know the things / data to express-!!

[...]

There are a couple of us on the forum focusing on NLP and context-based parsing (including myself). I think development in this area is key for better chatbot development, though you’ll find members here with very differing attitudes on the importance of language to AI in general.

However I think there’s a lot of room for projects like this one—figuring out the simplest mechanisms necessary to get a desired functionality. It’s very much an “engineering” approach to chatbot development. Coming up with a viable system for handling a specific domain of problems won’t necessarily require all the tools you’re referencing and it’s an interesting question in itself to ask just what is required. This project isn’t Chuck’s main chatbot endeavor, and at the very least, playing around can be fun and educational.

Hi Everyone,

@Andy, I appreciate your comments and suggestions.  I’ve managed to touch on most of what you’ve written, though admittedly not as deep as time permits.

I never have liked the expression “no need to re-invent the wheel”. It implies that the current solution is perfect or acceptable.  However, I do understand your intent. I do use DirectX for graphics and FMOD for sound in games I’ve created….no need to go out and create my own libraries.

By the way, there’s nothing wrong with your tone.  I’m somewhat thick skinned…and some times a bit thick headed…so what you wrote is okay with me. =)

@Dave and CR,
Thanks!

*Update*
Spent 8 hours last week working on programming the concept ‘confirming existence’ of something in the bots relative space (no specific XYZ coordinates).  The bot processes correctly several responses to single enquiries.  E.g. Is there a dog?  Are there lamps?

XBot can even process questions such as “Are there cats, dogs, and horses?” if they all exist or do not exist.  I’m now sorting out the process in which there are cats and horses, but no dogs.  I’ve developed a technique for managing singular and plural forms of nouns…it seems to work quite well.

This continues to be fun, time consuming, and frustrating. =)
Regards,
Chuck

Chuck

Whish you luck! (and good code-lines) I’ll give you an advice (hope it works) to manage those stuff, you are trying to make something like a query-language interpreter, normally this is accomplished by storing data in any SQL database and making a transfromation of the parsed query into an acceptable SQL query, then make some NLG to give a good shape to the outcome. Again if you manage this alone, even in clever code w/classes & many good relations you’ll reinvent the wheel (may be this time a better wheel than anyone.. but it will take you loooots of time) but belief me, the SQL engines out there, do work and make a terrific logical-optimized job. Even you can use some GNU and free-embedded ones, which manage all tables in memory and persist on a single file, they are very fast and worth looking at..
-best regards.
Andres

Chuck,
did you end up finishing a prototype spatial Excel app similar to your equation solver?