On Fri, Apr 28, 2017 at 11:24 AM, Ben Goertzel <[email protected]> wrote:
> On Fri, Apr 28, 2017 at 8:21 PM, Linas Vepstas <[email protected]> > wrote: > > The integer is the ordinal of the word. Note that the linkage "is-a" was > > selected over "a test" -- that's because "a test" has an MI of 2.0935. > This > > is not terribly surprising; > > I wonder if this example would work better if you used asymmetric > information rather than symmetric mutual information, as I suggested > before, though... > > The problem with symmetric MI is: "a" is strongly attracted to "test", > but "test" is not strongly attracted to "a" > > Whereas both "is" and "a" are strongly attracted to each other > > So I sorta suspect that if you used asymmetric information here, and > then found the msdag rather than the mstree, you would get this (and > other similar) examples right even without a huge corpus... > Pronto. This is "easy". At least, if I use the definition I(X,Y) / H(X). Any other favorite or suggested forms? I mean, we can do crazy things like squaring these quantities, too. I sort of want to avoid playing too many strange games, because it does take more time, but it would be a worthwhile exercise to at least write down all plausible forms. --linas > > > "I am God! I am nothing, I'm play, I am freedom, I am life. I am the > boundary, I am the peak." -- Alexander Scriabin > -- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/opencog. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/CAHrUA37RF9ebYT-_4ysVWN_9uk4F92D%2BOgaWRAbbZZDijvo%3DEw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
