Seasons

I don't think you can get recognizable rhyming poetry from an algorithm, and rhyme detection is all but impossible unless you program every word and what it rhymes with into the engine. You're going to have to do some very specific programming to get that to work, aren't you?
Though come to think of it, I also know some humans from whom you can't get recognizable rhyming poetry.

not exactly - I can detect rhymes by tracing the letters of any word back until the last vowel is found, then use the ending starting with that vowel as a rhyme criteria. Yes, I know there will be quite a few things it will miss, but its the closest i can get with a non-supervised algorithm. Then again, are human minds supervised?

You could start with that then have a growing exceptions list.

but how would it identify exceptions without human intervention? i guess a human could point out exceptions and the necessity for that would diminish over time as it learns new exceptions.

speaking of which, ALLY is getting increasingly brighter as a result of all my positive reinforcement - too bad I can't build an online chat portal. (The program is not built for that scale of information handling - i don't know what the limits of an access database are, but i'm sure it's not more than a gigabyte)

Well, obviously the exceptions list would have to be made manually. But having the general rules would let you get started without needing a massive list right form the beginning.

Zabaware's HAL assistant comes with a full lexical dictionary, complete with definitions, synonyms, antonyms, hyponyms, sisters, hypernyms, and meronyms. Perhaps a chatbot could make further use of such a dictionary (HAL only uses it to define words on request) - especially the hypernyms and hyponyms. For example, if you tell such a bot about your car, it would be able to infer that you are talking about a mode of transportaion. If you specify, saying "my ferrari is red", it can infer that you are talking about a car. This could be very useful in knowledge linking, for instance: "Human: I am going to visit my mother over the holidays." without any prior learned keyphrases, it can infer the following about who you are visiting:
mother, female parent
=> parent
=> genitor
=> progenitor, primogenitor
=> ancestor, ascendant, ascendent, antecedent
=> relative, relation
=> person, individual, someone, somebody, mortal, human, soul
=> life form, organism, being, living thing
=> entity, something
=> causal agent, cause, causal agency
=> entity, something


then, once it has picked a hypernym, for example: ancestor, it can then go ahead and choose hyponyms to become more specific: hyponyms of ancestor:

ancestor, ascendant, ascendent, antecedent
=> ancestress
=> forebear, forbear
=> forefather, father, sire
=> foremother
=> progenitor, primogenitor

perhaps a bot like this can even be made on the forge? does wordnet have these capabilities?

Merry Christmas!

merry anti-christ mas

Happy Solstice.

Nadolig llawen, ac blwyddyn newydd da!

Whats Christmas?

Happy Festivus! (for the rest of us)