On Fri, Sep 30, 2011 at 1:38 AM, John F. Sowa <sowa@xxxxxxxxxxx> wrote:
> As I said in my previous note, any meaning representation must be
> translatable to ordinary language, as a word, phrase, or sentence.
> Logic-based representations (including conceptual graphs) meet
> those requirements, but statistical vectors don't. (02)
I'm not sure what you mean by "statistical vectors". I use vectors of
word contexts. I think interpretations of these vectors must be
subjective, but are perfectly precise. (03)
I grant you quantum variables imply a certain randomness. Though I
think that randomness is actually just subjectivity too. (04)
In that sense your infinite lattice of theories is also random. It is
random in the sense that you can't universally identify any one theory
with "truth". (05)
I see no reason why it should not be possible to fold a sequence of
words out from a vector representing their combination. The process
strikes me as rather similar to identifying keywords associated with a
given search result in a search engine. Not totally deterministic,
perhaps, but perfectly feasible and practical. Search engines don't
currently index their search terms compositionally, of course, so
their mappings are still somewhat dumber than they could be. In
particular they don't index compositional structure. (06)
> ... very accurate mappings...
> ...accurate mental models.
> ...extremely precise mental models...
> ...The results are extremely precise.
> Similar examples of precision... (07)
On the question of precision. Distributed models are very precise.
They are also extremely robust. It's a feature. (08)
It is simply nice that they also provide a very principled way of
dealing with subjectivity. You might say they explain subjectivity,
why things appear to be subjectively "true". Indeed, why things can
often only be subjectively "true". (09)
An infinite lattice also implies subjectivity. I don't think even you
would claim it explains subjectivity though. (010)
> In every one of those activities, experts in the field communicate
> precisely with other experts. They don't depend on any kind of
> probabilistic mush. What they say is precise, and they can correct
> any mistaken interpretations to an arbitrarily precise level of
> accuracy. (011)
As an expert in this field, what do you mean by the term "probabilistic mush"? (012)
In general what loading are you putting on the term "statistical"? (013)
> most of the currently popular statistical methods in computational
> linguistics. ...
> The statistical people...
> We do use statistics... (014)
As far as I can tell all your uses of the word above refer to
different things. But that's my guess. Can you identify them? Who are
"the statistical people"? (015)
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (017)