Jeff, I see a methodological problem here.
CEPT representation of the word increases the size of the code with semantic information. This increases the size of the text presentation. The semantic information is not required in the task of remembering. This information will overload your system. On the other hand, you increase the words intersection. This strongly complicates the semantic analysis too. Do not get me wrong. I want to save people time and computers time only. Please do not explain if I'm wrong. Ivan
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
